"buy twice" is one of those live advices I heard and adhere to since.
It's basically the optimistic interpretation of "buy cheap, buy twice":
When I consider getting into something I buy cheap first, the idea being that that is enough to get a feel ...
... then you buy the second time and don't cheap out. But this purchase is more informed and you really get to appreciate it more because you know the step up from the cheap thing.
And sometimes... maybe even most of the time... the cheap thing is just enough.
As opposed to the old mechanics koan "Buy once cry once"
But sure, in general if you don't know what you are doing start cheap and treat it like a learning opportunity.
The real danger is in when you use shitty tools and think that because of that you are bad at something, so give up.
My personal story on this was when I hosed a laptop(don't disable usb when there are only usb devices) All the normal ways to fix it were not working so I went way out of my comfort zone and was going to try and reflash it the hard way with a chip programmer(It's already a brick I am not going to break it further) And I bought the cheapest sioc flashing kit I could find. and... nothing was working right, and because I have no idea what I am doing, does it just not work? do I have the wrong programmer? should I desolder the chip from the board? No clue. So it sat on the healing bench for a year. Then I stumbled on a forum thread complaining about cheap sioc clips, gathered my courage and bought a nice clip, tried again and it worked the first time. So on one level thanks Pomona electronics, your sioc clip was amazing, but the bigger lesson, I thought the task was just too unknowable and really I just had bad tools.
and think that because of that you are bad at something, so give up.
If you are learning a meaningfully new domain, you are bad at it.
If that causes you to give up, you are giving up on learning.
And in the case of your repair, the fact that the broken system sat on the bench for a year while you learned more meant that you did not give up.
Meaningful adult learning of new domains takes years. If something doesn't it either is not a new domain or the standard of performance is not an adult standard.
Finally, you could have bought the more expensive programmer first, but you might not have had the skill and knowledge and relevant experience to use it properly a year earlier...or to put it another way, you didn't have the direct experience to recognize the value of the post on which tool to use until after you had direct experience.
Buying tools twice or three or four times is just "the cost of doing business." You can't make an informed decision until you have experience.
Yes and buying the expensive item is not insurance against losing it, or breaking it, or it not being particularly useful. That’s before the possibility of buying the expensive version used…or the expensive item not being more robust.
On top of that, the canon of fine objects includes careful use and high maintenance…I know a guy who took offense at the suggestion he could drive his 911s in the rain.
And of course, you can buy a $45 keyboard twice and have a backup or one for your other computer, etc. Likewise, you can replace a $45 keyboard at 7pm on Tuesday at your local Walmart.
One of my student jobs was to transport film spools to theaters. They would arrive at my door in a box, I would walk them to the cinema on a small trolley and spend 2-3 hours in the projection rooms. The reels were spliced on site by a technician, projected, cut again and I transported them back home where they would be picked up again.
The job was less to transport the spools, but to supervise that there was no copying happening.
This was late 200x-ish, before digital protection became widespread iirc.
Actually yes. It's been a while, but now that you mention that, that probably was the reason for somebody (me) bringing the reels to the cinemas for a single showing in the first place.
"Back in the day" people were afraid that pupils would create CS (beta 6.5) maps of their schools. Gaussian Splatting would have been very convenient for that :-)
Yeah... We had those bulky TI Voyage 200 graphical calculators in school [1]. They could do everything the teachers could throw at us up to the point of having all but a few formulas build in.
I would say that definitely shaped me in a way where I rarely bother with the underlying details and tend to focus on how high level abstractions interact. [2]
[1] German "Mathe-LK", we could chose specializing in two things, for me it was math and computer science, the later being quite novel back in 2003.
[2] I _do_ tend to specialize in things, but e.g. for LLMs or GLMMs, while I do have the capability to understand the technical details, I just don't bother.
I am always a bit baffled why Apple gets credited with this. Unified memory has been a thing for decades. I can still load the biggest models on my 10th gen Intel Core CPU and the integrated GPU can run inference.
The difference being that modern integrated GPU are just that much faster and can run inference at tolerable speeds.
(Plus NPUs being a thing now, but that also started much earlier. Thr 10th gen Intel Core architecture already had instructions to deal with "AI" workloads... just very preliminary)
That’s shared, not unified, it’s partitioned where cpu and gpu copies are managed by driver. Lunar lake (2024) is getting closer but still not as tightly integrated as apple and capped to 32GB only (Apple has up to 512GB). AMD ryzen ai max is closer to Apple but still 3 times slower memory.
Shared vs unified is merely a driver implementation detail. Regardless, in practice (IIUC) data is still going to be copied if you perform a transfer using a graphics API because the driver has no way of knowing what the host might do with the pointed-to memory after the transfer.
If you make use of host pointers and run on an iGPU no copy will take place.
My last serious GPU programming was with OpenCL. And if my memory does not fail me the API was quite specific about copying and/or sharing memory on a shared memory system.
I am pretty sure that my old 10th gen CPU/GPU combo has the ability to use the "unified"/zero-copy access mode for the GPU.
I don't think people are crediting Apple with inventing unified memory - I certainly did not. There have been similar systems for decades. What Apple did is popularize this with widely available hardware with GPUs that don't totally suck for inference in combination with RAM that has decent speed at an affordable price. You either had iGPUs which were slow (plus not exactly the fastest DDR memory) but at least sitting on the same die or you had fast dGPUs which had their own limited amount of VRAM. So the choice was between direct memory access but not powerfull or powerfull but strangled by having to go through the PCIE subsystem to access RAM.
The article is talking about one particular optimization that one can implement with Apple Silicon and I at least wasn't aware that it is now possible to do so from WebAssembly - so to completely dismiss it as if it had nothing to do with Apple Silicon is imho not fair.
Can't confirm. We had students at university (18-20-ish) that had not used a mouse prior to our courses. That was at least 3-4 years ago now and not a single case.
So something that you can do with PDKs is add your own custom standard cell and tell the EDA tools to use them. This is actually pretty smart, this way you can use most of the foundry cells (which have been extensively validated) and focus on things like this "magic multiplier", that you will have to manually validate. This also makes porting across tech nodes easier if you manage only a handful of custom cells versus a completely custom design.
(I have my guesses as to what that is, but I admittedly don't know enough about that particular part of the field to give anything but a guess).
My "only" experience here is designing ASICs for Neuromorphic Chips. We used sub-threshold exclusively for linearity and energy reduction. No standard cells for us
We have been running Ardour 9 for a while now during band rehearsals. Currently 12 channels that we record and monitor in realtime with some effects on top.
I am trying something similar: basically a LLM managed MUD for people with too little time for roleplaying.
If you want to chat, hit me up at mail-from-ohsohumble@f12n.de
reply