I have a gut feeling, next in line will be 2 or more level of MoE. Further reducing the memory bandwidth and compute requirements. So top level MoE router decides which sub MoE to route.
The way I see it, it is more like a very-high initial negotiation position. From here on, each country will be dealt with individually. You accept our terms and get 5% discount on tariff ! Every country has a set of red lines. Fewer the red lines better they will be placed in one-to-one "trade agreements".
EDIT: Someone from white-house explicitly declined tariff to be a negotiation. Next few days will be interesting.
Seems it is a story time thread. Here goes my strangest one.
Back in 2005, when I had only paid-by-cash internet cafe access to computer, one of the shopkeeper offered me free time on computer IF I typed and ran a 15 page of class 12 computer project printed on A4 sheets, onto the compiler. TurboC++. I gladly accepted the offer and typed things.
When I finished typing, taking out all the compile error, the program didn't work as expected. Few hours latter, I find out that 1 or 2 pages of printed source codes were not in original order. :-O . So had to swap code from one function to another to finally get it working. That was one hell of a lesson!
Shopkeeper must have sold that project to many students, and I got some Free internet access.
Thank you! We're probably going to have to eventually use every engineering software in existence haha for CAD, mainly solidworks and CATIA, for EM Physics modeling: COMSOL. But most of the analysis is done with specialized plasma physics and radiation transport codes like MCNP or Open MC.
How much would the Non-Recurrent Cost associated with designing a RISC-V CPU using this 2nm? EDA Tools, Photo-Masks, One Time Chip Design Engineer Cost, Simulation/Virtual Verification and so on. I mean every thing till tape-out.
Riscv doesn't need smaller nm. It just needs someone to actually design and release a good core design. 14nm (or 4 or anything in between) is perfectly suffient to make a Riscv chip 10x faster than any Riscv that currently exist.
14nm or so is kind of a sweet spot for general purpose chip design right now, because later nodes turn out to have higher overall per-transistor cost despite the improvement in density and area. Of course this may well change over time as even finer production nodes get developed and the existing nodes then move closer to the trailing edge.
The article linked in another comment pointed out that cost per transistor keeps falling, and it's just the fast increasing fixed costs that make it seem otherwise.
Interesting! Could you perhaps point me towards the source where I could read up on the state of the art of chip manufacturing and the implications coming from the respective manufacture processes?
There are already companies making huge, expensive, high-performance chips for x86 and ARM; chips which are operating at the limit of what's physically possible. There aren't chips which do that for RISC-V to my knowledge.
I don't see how that changes things. The end goal is the same. x86 and Arm are much further along, but that doesn't mean they "need" better lithography any more or less than RISC-V "needs" it.
Not a hardware person but I read in an interview with Jim Keller that ISA itself doesn't matter that much for performance.
>[Arguing about instruction sets] is a very sad story. It's not even a couple of dozen [op-codes] - 80% of core execution is only six instructions - you know, load, store, add, subtract, compare and branch. With those you have pretty much covered it. If you're writing in Perl or something, maybe call and return are more important than compare and branch. But instruction sets only matter a little bit - you can lose 10%, or 20%, [of performance] because you're missing instructions.[0]
I've cited this article a few times already (and seen others cite it) so if this is incorrect I hope someone could correct me here. (I have to assume that you also need some sort of SIMD/Vector these days, which RVA23 has anyway, but aside from that.)
I've also read that you can port CPU cores to a different ISAs pretty easily which is what PA Semi did when Apple bought them (M1 devs). So what seems to be missing is for a bunch of senior CPU developers who worked at AMD, Intel, PA Semi/Apple or ARM to just make one for RISC-V. Not sure if that is what you meant by IP here. Tenstorrent could be one such group, and they are building RISC-V CPUs, but their focus seems to be split between that and AI accelerators. China is another good candidate.
By IP I meant intellectual property. I was wondering how much designs or specific hardware developments that were essential for the performance boost (or: energy efficiency of chips, compute/W) as observed with the Apple M family of designs are locked down by patents.
Keller is just a god tier hardware guy - his ability to leverage deep understanding and then explain incredibly complex issues in a few sentences is incredible. Intel shot themselves in the foot when they let him go.
One of the things that could really help here is to create a license along the lines of "GPL for hardware" and then for a company like Google or Meta to release a design under it. The design wouldn't even have to be state of the art; something with performance equivalent to a ten year old x64 design would be valuable enough that people would use it for things if it was free.
But the larger value is that then people could use it as a starting point for modifications, which would in turn have to be released under the same license. Soon you have a diverse set of designs to choose from and we can start getting open hardware into common use because using any of those designs would be a cost savings over buying or designing something equivalent, but because the license requires an open design the user can then create custom firmware etc. and we might finally start to do something about the IoT security nightmare.
The RISC-V core designs that have emerged so far don't have nearly the amount of silicon dedicated to optimizing IPC as one sees in ARM, x86 and others: predictors, sophisticated caches, complex instruction dispatch, deep pipelines, etc. That stuff isn't provided with the RISC-V core designs you get for free or license at low cost, because it's fabulously expensive to develop, tied to the ISA and core designs for which it is created, and jealously guarded through IP law.
Will RISC-V get there eventually? People like Jim Keller are building companies around that goal. However, it will likely take years, at least, for RISC-V to approach parity.
Only reasons it wouldn't all boil down to, `it wouldn't matter`. E.G. If something else that does have it gets released to free as in beer levels why bother?
Unless modern civilization ends or an easier alternative is released eventually anything Open Source will get good enough.
Building a chip is really hard, and the companies that have experience all have their own architecture that they like and no reason to make a high performance design that competes with their regular chips.
I am a civil-structural engineering and have obvious bias for Concrete. Over here in India, literally all houses are built using Concrete and Burnt-Clay-Brick / Fly-Ash-Brick Masonry. I hope Concrete gets promoted more as a building material. They buildings which are professionally designed easily withstand 2500 Year Return Period magnitude earthquakes. Last time I enquired on HN about preference for Wood in US (remote areas) Building Materials, someone said, can't design house venerable to High Seismic Activity. While my exposure to US Building codes is limited, I know for sure, ASCE (American Society of Civil Engineers) have excellent Earthquake Loading criteria. It should be doable. Perhaps some regulation could help there.
A personal example: Wardrobes are usually made using synthetic-wood over here in India. I went a step ahead, and got it build using Steel-Sheets. So a major chunk of fire potential removed from the house. And it was within 10% of the cost of wood work. Termite free forever as bonus ! Have a look at the photos in google maps listing of the local manufacturer. https://maps.app.goo.gl/7Wrt4rNtcpez53Bm6
Wood is incredibly cheap here in the United States. Estimates I have seen in the past is a stone home will cost 15-25% more per square foot than a wood-framed home in the same location. Making it more resistant to earthquakes (a requirement in California) raises the price even further. At the end of the day, cost will almost always win.
One thing I think about is concrete is a major consumer of energy and contributes a large amount of CO2, something like 8% of the global emission. Whereas wood literally grows on trees while sequestrating said carbon. I realize that there re efforts to make concrete carbon neutral but until that happens building with concrete is not environmentally friendly.
But if you live in a tinderbox like California, that will only get worse from climate change, how long will that sequestration actually last? How much carbon will be emitted rebuilding and replacing?
Those people need to be housed either way so that energy is already spent. The path forward for Californians or any one in fire prone areas is to practice realistic fire-resistant home building and landscaping. Many of these techniques developed do not require concrete. That or these areas will have to be no-build zones.
Adding to that, concrete buildings rely on a lot of steel, which is another major emissions driver, and production of both also leave a lot of toxic waste behind.
Construction should try to use both sustainable materials wherever feasible, and strongly favour refurnishing existing houses over new buildings.
We also build with concrete. In the USA wood is very cheap and it's easier to work with, so you end up with a larger house at half the price of a concrete house. Also wooden houses have a "warmth" that is missing from bricks.
The way Americans look at it is: I can get a house twice as big for less. I'll just get insurance with the money I save.
Houses almost all use wood framing. Rarely they use steel framing, which is more expensive and provides worse insulation. None use concrete or masonry because it is illegal to build a house that will collapse during a M7-8 earthquake. Like Japan, construction style in the western US is driven primarily by the requirement to be extremely seismic resistant, since that is a predictable and unavoidable risk.
In Southern California, it is typical to have tile roofs and stucco exteriors, which helps protect against the embers that will rain on your house during a major wildfire.
Most houses in the US are made of wood. In Los Angeles they often have tile/concrete roofs, but I've read that in a situation like this the problem is the vents under the edge of the roof that lead into the attic: if anything burning gets through there, the house is toast.
Source: used to live in a Los Angeles hilly suburb. If the fires get to where I used to live, that house will definitely burn despite having a cement tile roof.
Insurers require ember resistant vulcan vents and the like now. It’s a relatively minor upgrade for most homeowners since its just a mesh over the vent.
No, vulcan vents only help when the exterior is capable of resisting the fire. These winds threw fist sized pieces of burning wood for hundreds of feet which is much harder to defend against.
All homeowners I know here have already had them installed over the last five years because of fire insurance inspections but I don’t know how representative my sample is.
Considering that winds both made the spread a lot further, faster and moving around fuel (trees/wood/material blowing around and ending up on streets and whatnot), it sounds unlikely any 1 solution would have prevented this.
The house was framed in 2x4 which is standard in most of the US. We did have modern fire resistant siding and roof tiles, but this was leagues beyond what any of those materials are designed for. It melted most of the steel around the property. Thank you for the kind words.
