I'm not surprised at the outcome. These Ampere system single core/thread performance is pretty low and that is where you feel it. The OS/software simply cannot allocate the threads across enough cores effectively to make up for this difference.
This is why things like the Apple M Series feels so fast, because while they don't win the multi core performance especially when going up against a 80 core beast like this, they have single thread performance exactly were it is needed.
Maybe we will need 80 cores in future, that is cool but for daily home use it is still just way too much for what we need.
Apple design their own Arm-compatible cores from scratch. Ampere use a modified Arm Neoverse N1 core. In addition, the Ampere server that Marcin is using is about 6 years old, and would have been tuned for core count over single thread performance (good for web serving). Basically Arm's own cores aren't nearly as good as Apple's at the best of times, and having a 6 year old server makes things even worse.
Ampere's primary focus is running lots of simple tasks concurrently, at relatively low power, with lots of I/O. So, many tens to hundreds of cores, not too fast, at lower power draw than amd64, with lots of PCIe lanes for storage and network.
Apple's primary focus is user experience and power efficiency. That's why you'll find a handful of fast performance cores and low power efficiency cores, along with graphics acceleration to drive high resolution displays.
To expand on what others have said, aarch64 merely defines the set of instructions that the CPU can perform. That has an impact on how you design processors, but you can still design multiple completely different processor architectures with different performance characteristics that all implement the aarch64 instruction set.
The whole thing just screams square-pegs-in-round-holes, for a desktop PC he bought a data-centre-server MB with a CPU with $ludicrous cores with an unsupported (qualified) GPU and a custom-built kernel... it sounds like he's trying to get a spot on Animarchy's YT channel, with his trademark line "And then... it got worse".
> I work at Red Hat. Mostly on AArch64 support in several projects.
So an ARM developer, working for a major Linux distro vendor and trying to dogfood their work, used the closest thing to an ARM workstation that Linux can run on.
What other alternatives would you suggest? The various Apple Silicon or Snapdragon laptops that have their own well-documented problems running Linux? A smartphone running as a desktop?
There aren't very many ARM-based options that are even feasible for use as a developer desktop, even if the software did work correctly.
Apart from the obvious "well, don't do that then", I'd probably use an ARM-based ODroid for development work, and if I needed more horsepower, the ODroid as a front-end for said thing with more horsepower.
(I actually do development on ODroids, they're quite nice, if underpowered compared to the Intel/AMD equivalents).
Very cool to see, so pointless it just had to be done.
I was wondering how it was squeezed into 64KB of RAM but it uses the 4MB on the Everdrive cart. With that it makes sense, considering Linux can run on an N64 with 4MB of RAM.
in terms of running linux, yeah it kinda is. Both archs are supported by the kernel, and the kernel doesn't need an MMU if you compile it for a system without one
I'm not denying it makes things easier. But if the question is a binary "Can it run linux?", the flowchart is pretty much "what kind of cpu does it have" and "how much ram can it use"
I ended up getting a second hand Optiplex Micro for this. Tiny unit, low power usable, never even heard the fan switch on. Even with the slow frequency (2ghz) the Intel media decoders are brilliant at handling this stuff.
Mine was built with a leftover ryzen 1500X, microatx motherboard and RAM that were effectively free, a geforce 1030, and a random cheap 256GB SSD I found on newegg (the video content lives elsewhere across the LAN). It continues to be capable of playing 2160p60 H.265/HEVC content so I don't see a cpu and motherboard upgrade any time in the next couple of years, unless very high bitrate AV1 encoded content suddenly becomes more popular.
That sounds like it should be good for many years to come. Maybe if there is an uptake in AV2 once that is out then issues will come up. I'm not a shill but the one thing Intel have done well with their ARC cards is the media decoder. If their GPU space doesn't work out at least they have that and that could be a decent upgrade path in future.
I've tested it with 'normal' bitrate 1080p and 4K AV1 content and it still keeps up, staying under 70% CPU usage on all four cores, all AV1 decode in VLC is done in software since the geforce 1030/1050/1070/1080 generation of cards obviously has no capability for AV1... We'll see how it goes in 3-4 years.
I mean, I guess I daily run an old computer. Lenovo T400 from 2009, 2.4Ghz Core 2 Duo with 4GB of RAM. So far I haven't really had any issues. That said today I picked up a Carbon X1 4th gen for $100, that might be become my new Old computer. Also in the process of refurbishing an IBM Aptiva from 1996. Pentium 166Mhz with 64MB RAM, that one is a little beauty.
I do like this years challenge, 'hand-make something' as that is always a good thing to do.
T400 is definitely cool, I have a T420 I love. But that Aptiva... man, when those were new, they were full of more bloatware than any other machine I'd ever seen. I made bank in high school just reinstalling windows for people on those... it was like 5x faster when I was done with it, easily. I think it's made it impossible for me to see that as a beauty, though.
A T400 is old enough to feel refreshingly simple but still new enough to be useful. A 1996 Aptiva is more like a historical instrument. I like that distinction between old and actually vintage.
A counter point is that if you are stuck in a 50/50 position, out sourcing the decision to some random other thing can break the dead lock. Like flipping to a random page on a book and going with the first word you see that is positive or negative (Bibliomancy).
The occult language can make it sound mysterious but it can also sometimes just be a cover for a delegation of decisions.
It's a weak counter point in that it does not consider the long term effects.
Using I Ching to make decisions is like using an LLM to think. At some point, you will atrophy the faculty of independent decision making.
You are also neglecting that I Ching outcomes come with text and commentary that are by definition designed to affect your mindset -- "the image" -- (for the good of course but see below for that). This is precisly what Dr. Jung was saying about trifling with the occult. It affects you at an sub/un-conscious level - this is not some random book. It is the I Ching, and like all world scripture it has a textual potency that affects its readers.
If you are stuck at a 50/50 position, just throw a single coin. Average 3 if you must. Spare yourself the commentaries.
Correct and effective use of the I Ching requires a degree of maturity and self development (think Carl Jung) that is surely lacking in most of us when we are first introduced to this occult artefact in 20th and 21st centuries. That is because in this age of facile information we get our hands on matter that in previous eras were obtained after spending years at the feet of some guru or master!
I don’t think “just throw a single coin” really captures how I Ching divination traditionally works.
The classical Dayan method is based on human participation through division and transformation. It is the only divination method recorded in the Book of Changes. Through repeated divisions, six yin or yang lines are formed, together forming one gua. To me, this feels closer to life: we make choices, exchange one thing for another, and move from one situation into the next.
Maybe the gua is simply a manifestation that helps us see the situation more clearly. Some people believe in it, and others do not. To me, its wisdom is a bit like a zero-knowledge proof: it reveals a pattern without fully exposing the mystery behind it.
It’s great to see people discussing this. I’m a huge fan of the I Ching and built a few websites around it, including https://knowunknowable.love and https://ichingdao.love, where I explore the I Ching, Daoism, and Mozi.
I didn't say that is how the I Ching works. Participation in the process is most certainly an important aspect of binding the psyche of the questioner to the results, which supports the view that the stalk method is indeed the way to consult the book, as that requires contemplating the question while going through the process.
That's fascinating. The I Ching is centuries old. It's survived this long. The concept that knowledge should be witheld from people until they're ready is so anthical to our current beliefs. Or at least mine. It's history says that's not an invalid way of thinking about the world, so I'll have to think on that.
> The concept that knowledge should be witheld from people until they're ready is so anthical to our current beliefs.
It is not knowledge that is witheld. It is text. I also propose to you to consider the thought that true knowledge is only obtained by experience. The fundamental issue with psychologically potent text and images is that it can induce convictions based on misreading and misunderstanding.
IFF you can exercise the necessary self control to avoid premature utility of the I Ching, the entire work is of course a profound metaphysical model of human reality through the lens of Taoist sages based on a simple and elegant axiomatic framework that uses the concept of state-transitions & numbers 2 (forces) and 3 (places) to develop a comprehensive metaphysical system to consider "all things under the heaven".
This perspective always fascinates me. I'm not trying to be negative, but am genuinely curious.
If you're truly split like that, why is any ceremony necessary? Just pick one. Does every choice have to be explained? Are you doing this to amuse yourself?
If you don't delegate to a source of randomness, do you feel guilty or ruminate about why you were un/successful? Are you afraid of your own thoughts misleading you once the outcome is clear?
This is the path that Japan tried to go down and it hasn't worked out yet, but we have also solved a lot more of the technical issues since they began. going to be interesting to see if we pull it off this time.
Humanoid robots barely progressed between 2000-2020. There have obviously been incremental improvements in things like dexterity, vision, self-balancing, and locomotion, but in terms of having a useful humanoid robot, Honda's ASIMO released in the year 2000 is not crazily behind what we had in 2020. So it's not surprise we haven't seen economic dividends yet in the real world.
I think AI is what could make humanoids turn from parlor tricks to huge amounts of utility, but we're really going to have to see how it plays out in the next 5-10 years.
I think they mostly tried to go down this path before we had the transformer. With VLA models, or really now "Large Behavior Models", what's possible has changed dramatically. I've seen robot arms fold laundry now. Textile work is insanely hard, now it's just putting a lot of learned behavior together.
The current humanoid hype don't have much substances or key technologies in it, and incumbent industrial robotics companies like FANUC are already in the process of rolling the techniques created for humanoids into their robots. I personally think this is going to be just series of incremental gains for big welding bots, and nursery equipment becoming mildly robotic, like Aperture Science wall panels, than humanoids walking into retirement homes and doing dishes in the future.
Those were never techniques created for humanoid robots. Google researched using transformers as vision language action models to drive robots back in 2023 on a mobile manipulator and probably did non VLA work even earlier.
This is something people refuse to understand. The shape of the robot changes absolutely nothing about robot intelligence unless it abandons the basic concept of joints and links. Continuum robots are very difficult to control but they are also incredibly niche.
I really can't shrug off my suspicion since one of Chinese humanoid startups had added few female type body bumps on a robot and the online humanoid hype cluster("but does it do laundry or dishes" types) responded in panic with shared sentiments of defeat, that, those humanoid hype types might be just yearning for HBO Westworld bots, to state it with modesty, not even Data or Marvin who we can share drinks with. The absolute humanoid obsession doesn't make sense to me otherwise. The cookie cutter "humanoids are best poised for leveraging pre-existing human infra" response make close to zero sense otherwise.
Dreamweaver was cool as a beginner because it took a lot of the troublesome parts out of the equation. But it did end up being more of a hindrance than a benefit the further you went in.
I never understood Dreamweaver. The first thing it asked me when making a new website was ... what the resolution of my user's screen is? I don't know that!
Its web development software from the 90s/00s, a period when websites were built by first having a designer meticulously mock everything up in Photoshop on a 640x480 canvas (maybe 800x600 or 1024x768 in later days), that mockup would then be handed over to a web developer (hi, that was me) who would take that mockup, slice it up into a billion little images, and then put them in a wildly complex set of nested HTML tables. The designer would then have a look over it and provide critique on the fact some element was 3px misaligned, or the font size was incorrect.
During this period I was berated by our studio lead for using new fangled technologies like CSS layout that could adapt to different sized screens instead of sticking to the trusty HTML soup Dreamweaver would spit out.
There was a ton of... not exactly footguns, just things to keep in mind if you’ve wanted your site to work as you intended in all browsers. The webcompat nowadays is way better now.
That said, personally I’ve never understood Dreamweaver either. By the time I tried it, I’ve already got used to Notepad++ and writing HTML by hand, so I’ve just treated it as another text editor... and IIRC it just felt way more laggy than Notepad++, with a browser preview panel that took half of my 4:3 display. Maybe I’d discover some cool features if I’ve spent some more time in it? I dunno.
> HTML at the time was intended as an application of SGML
Worse, it was an extended superset (ha!) of SGML. At least 20 years ago, SGML::Parser would reject some valid HTML documents.
That said, it was really easy to type correctly in a text editor (especially compared to actual SGML), particularly one that indented and matched tags for you.
Just like AI vibecoded websites... Good luck understanding the code when the AI bubble explodes and you can't afford the insane price that AI will have by then.
I have wondered that. I have this one and a very VERY bland Facebook that I post something stupid on every 12 months or so. So in a strange way I have accidentally shielded myself.
The argument usually is that it is a slippery slope. Something that is introduced in the name of virtue ends up being co-opted into a system of control as those in power and peoples attitudes change with subsequent layers of normalization.
Sure, maybe that's true. Who knows, maybe it will also actually just work and solve the problem. There's no argument there.
The argument that touches on "those in power" must also contend with "those in power" just abusing that power anyway, no matter what legislation is passed. People in power are usually pretty crafty.
This is why things like the Apple M Series feels so fast, because while they don't win the multi core performance especially when going up against a 80 core beast like this, they have single thread performance exactly were it is needed.
Maybe we will need 80 cores in future, that is cool but for daily home use it is still just way too much for what we need.