|
Cygni posted:I guess this would be the right place for Zhaoxin discussion even though its VIA's x86 license? you guess the explicitly non-x86 platforms thread is the right place for x86 news?!
|
#
?
Dec 15, 2023 18:08
|
|
|
|
| # ? Apr 29, 2024 02:02 |
|
|
Farmer Crack-rear end posted:you guess the explicitly non-x86 platforms thread is the right place for x86 news?! yeah
|
|
#
?
Dec 15, 2023 18:13
|
|
|
which thread do i post about XScale in??
|
|
#
?
Dec 15, 2023 18:14
|
|
|
repiv posted:which thread do i post about XScale in?? Create something in Pet Island I guess.
|
|
#
?
Dec 15, 2023 18:16
|
|
|
Did HP ever release ARM blades (ok, cartridges) for Moonshot? I remember that they were pitching heterogenous compute architectures in it way back when. I guess this is around when AMD tried to make ARM chips too!
|
|
#
?
Dec 15, 2023 18:20
|
|
|
Twerk from Home posted:Did HP ever release ARM blades (ok, cartridges) for Moonshot? I remember that they were pitching heterogenous compute architectures in it way back when. They were around, yeah. The cores in them were pretty unimpressive.
|
|
#
?
Dec 15, 2023 18:57
|
|
|
HPE has a couple arm options right now; if you want to spend a lot of money you can get the rebadged Fujitsu Fugaku nodes at $$$. They also have Ampere at $ but I only saw the 80 core skus when I looked recently.
|
|
#
?
Dec 15, 2023 19:00
|
|
|
I never heard about or encountered moonshot as anything other than the butt of a joke, did anyone actually buy them?
|
|
#
?
Dec 15, 2023 19:21
|
|
|
There’s some x86 ones on eBay, so apparently?
|
|
#
?
Dec 15, 2023 20:06
|
|
|
Anyway moonshot was a lot more credible than The Machine https://en.m.wikipedia.org/wiki/The_Machine_(computer_architecture) Actually an interesting architecture but it was built in anticipation of tech that never* made it to market and a new OS was a big ask, too. * ok persistent ram got to market with Optane but we all know how that worked out and Intel wasn’t going to put it on non x86 anyway.
|
|
#
?
Dec 15, 2023 20:26
|
|
|
now who would be mad enough to try to stick an Ampere Altra in a desktop form factor… https://www.newegg.com/p/N82E16813140134
|
|
#
?
Dec 22, 2023 07:55
|
|
|
Q64-22 is only 69W according to Ampere, so it'd do fine in an ATX tower. I'm guessing this board exists because people working on software to deploy to arm64 cloud servers need development & test workstations.
|
|
#
?
Dec 22, 2023 08:20
|
|
|
if only ampere or someone would make a small chip with the same systemready certification so we can finally have SBCs with good out of the box distro support
|
|
#
?
Dec 22, 2023 13:15
|
|
|
RISC-V based "gaming console" https://www.tomshardware.com/pc-components/cpus/risc-v-handheld-gaming-system-announced-linux-as-the-basis-for-a-retro-gaming-platform There's a huge cottage industry of ARM based steam deck looking devices that come preloaded with a bunch of emulators like Sega Dreamcast, SNES, PlayStation 1/2 and hundreds of  romsI guess this is the "if you build it they will come" moment. Now all the retro emulators have an excuse to add RISC-V as a build target Looks like price point around $250 USD, maybe
|
|
#
?
Dec 23, 2023 18:07
|
|
|
i was checking out duckstation the other day and noticed it already has a RISC-V JIT backed, not sure what existing hardware that's aimed at
|
|
#
?
Dec 23, 2023 18:15
|
|
|
RISC-V is almost exclusively supported by Linux, so it's (probably) just a RISC-V Linux binary and talks to the kernel for audio video bindings. Presumably whatever custom kernel was compiled for the device should have adequate GPU support and whatnot. Ubuntu has had official support for RISC-V since the April 2020 release so presumably by the time this ships RISC-V will have been mainline for 4 years I'm increasingly of the opinion that the US government missed the boat on containing Chinas CPU capacity by about five years
|
|
#
?
Dec 23, 2023 19:36
|
|
|
Hadlock posted:RISC-V is almost exclusively supported by Linux, so it's (probably) just a RISC-V Linux binary and talks to the kernel for audio video bindings. Presumably whatever custom kernel was compiled for the device should have adequate GPU support and whatnot. Ubuntu has had official support for RISC-V since the April 2020 release so presumably by the time this ships RISC-V will have been mainline for 4 years I'm increasingly of the opinion that you have a weird boomer-like obsession with China, including Clancy-esque ideas about somehow preventing them from developing domestic CPU design capability. Also that you just don't understand how CPU design works, which ties in with actually believing the clancy nonsense. RISC-V hasn't helped China at all. It's not even a very good ISA, but more importantly, ISA is the least important thing here. x86 may be sitting a bit uneasy on its throne, but for now it still rules the PC world despite being an objectively bad ISA design. What matters is putting together a team of talented and experienced design engineers who can make a high quality, low power, and fast implementation of any ISA. The experience part of that takes iteration (AKA practice), same as anywhere else in the world. There is no way to stop China from practicing. Even if it was necessary for them to practice with the best commercially relevant ISA (it's not), they could always just buy an Arm architectural license. By the time they have one or more world class implementation teams, they can worry about designing their own RISC ISA, and retarget for that. Of course when I say "they" about China, I'm falling victim to the weird narrative you're forcing on things, in which China is a unitary executive. What's actually going on is that China is a capitalist state. Yes, it has an authoritarian single-party government that kinda pretends to still be communist, but it's got a ton of very rich people who privately own a lot of the means of production. Some of these capitalists are trying to break into CPUs. They often get some support from the Chinese government, in fact some of them seem to be entirely propped up by it (something we do for strategic industries in the West too), but ultimately they are capitalists. They want to make chips which people will buy. Lots of the market for CPUs is outside of China. Hence the interest in an up-and-coming ISA which might appeal to the West, yet does not require paying licensing fees or unit royalties to Western companies. That's it, that's the main reason Chinese companies like RISC-V. It's just markets and money. As far as I know there's no Chinese company with a world-class ISA design or CPU implementation team yet, but it's a fool's errand to try to stop it from happening.
|
|
#
?
Dec 24, 2023 14:17
|
|
|
BobHoward posted:That's it, that's the main reason Chinese companies like RISC-V. It's just markets and money. As far as I know there's no Chinese company with a world-class ISA design or CPU implementation team yet, but it's a fool's errand to try to stop it from happening. Well, we are trying anyway: https://www.reuters.com/technology/export-controls-hit-chinas-access-arms-chip-designs-ft-2022-12-14/. https://www.theregister.com/2023/08/23/china_arm_tradewar/ Neoverse V2 is already illegal to license to a Chinese company, and ARM is scared that they'll be blocked from licensing other designs in the future. I'm not Hadlock but it's super easy to buy into a "Cold War 2" mindset with all the coverage around this. It's also not like we can restrict any ARM licensing anyway, if a regulation changes then whoever used to be making them will just keep making them without a license. I guess it would dramatically slow down the introduction of new ARM reference designs.
|
|
#
?
Dec 24, 2023 15:41
|
|
|
BobHoward posted:RISC-V hasn't helped China at all. It's not even a very good ISA, but more importantly, ISA is the least important thing here. x86 may be sitting a bit uneasy on its throne, but for now it still rules the PC world despite being an objectively bad ISA design. What does and does not contribute to a good ISA design? Is it the inevitable fate of any sufficiently old ISA to accumulate enough cruft to bog it down?
|
|
#
?
Dec 24, 2023 17:20
|
|
|
Twerk from Home posted:Well, we are trying anyway: https://www.reuters.com/technology/export-controls-hit-chinas-access-arms-chip-designs-ft-2022-12-14/.
|
|
#
?
Dec 24, 2023 17:44
|
|
|
BobHoward posted:What matters is putting together a team of talented and experienced design engineers who can make a high quality, low power, and fast implementation of any ISA. The experience part of that takes iteration (AKA practice), same as anywhere else in the world. I always think about the kind of folks who can tie specific CPU architecture features to things like how many layers the PCB needs. Sure you can distribute that decision into a team or 5, but.. I've met individual people who could do that kind of thing no sweat. In college we had an assignment that was optimizing a core, deciding where to spend transistors. 4 way 8 set iCache or spend those on the BTB. Apparently enough students complained about it that the prof took 10 minutes to explain why "design an ISA" is a bad assignment. Nobody gets to do that professionally (without decades of experience like Shippy), it's awful to grade, etc. etc.
|
|
#
?
Dec 24, 2023 20:16
|
|
|
JawnV6 posted:I always think about the kind of folks who can tie specific CPU architecture features to things like how many layers the PCB needs. Sure you can distribute that decision into a team or 5, but.. I've met individual people who could do that kind of thing no sweat. Specialization is very important in certain areas but "systems thinking" is a common trait in the highest-performing people I've worked with. If you can connect a narrative thread behind your CPU uarch / silicon designs to the package impacts to the implementation requirements on the PCB in your brain, that's a superpower -- don't need a team of systems engineers managing requirements. "This SKU must be low cost -> package constraints -> needs to be implementable on 4L PCBs at the cheapest fab imaginable -> constrain to 0.8 mm pitch BGA -> pick FCBGA -> can't do insane amounts of power or high-speed I/O given those constraints -> etc" Chiplets might be one of the best things that's ever happened for being able to play LEGO on the chip package itself -- and everyone is getting better at the innovation / development required to make that a good idea (inter-chiplet comms / etc).
|
|
#
?
Dec 24, 2023 21:03
|
|
|
ConanTheLibrarian posted:What does and does not contribute to a good ISA design? Is it the inevitable fate of any sufficiently old ISA to accumulate enough cruft to bog it down? It doesn't have to be inevitable, a good steward can avoid putting dumb things in. x86 didn't need decades of cruft accumulation, though. It had bad things from the word go. Some of the original design flaws kinda got shoved off into a corner by the 386, so long as you ran only 32-bit software. (Something which took more than the market lifespan of the 386 to actually achieve since Microsoft was so incredibly slow to fully take advantage of the 386.) Others are still with us today.
|
|
#
?
Dec 24, 2023 23:25
|
|
|
Twerk from Home posted:Well, we are trying anyway: https://www.reuters.com/technology/export-controls-hit-chinas-access-arms-chip-designs-ft-2022-12-14/. Yeah my dad worked at Intel as an engineer for like, 3-5 years while i was a kid so I'm probably ever so slightly more attuned to processor stuff in the news than your average autist. And the news feed algorithm is going to feed me a steady drip of news on the topic. And this is the RISC-V thread, so And yeah the Biden administration has been making statements and actively locking down CPU technology share for... At least the last year? I think he's made public statements every week for the last two months. We went from 0-100 on CPU policy lock down recently it seems topical. Members of Congress (2?) have made public statements about shutting down RISC-V knowledge transfer recently, which is pretty weird Outside of the recent change in policy I have not a lot of interest in China. Although I guess full disclosure I've been to Hong Kong like, three times, but I travel a medium amount in the region by American standards Edit: oh and I'm writing some software that's targeting RISC-V specifically
|
|
#
?
Dec 24, 2023 23:26
|
|
|
BobHoward posted:There is no way to stop China from practicing. Even if it was necessary for them to practice with the best commercially relevant ISA (it's not), they could always just buy an Arm architectural license. they don’t even need to do that, they could just implement the instruction set, since that’s not protectable IP in and of itself of course if they wanted the best ISA they’d be doing a 64-bit extension of the Motorola 68000…
|
|
#
?
Dec 25, 2023 01:03
|
|
|
eschaton posted:they don’t even need to do that, they could just implement the instruction set, since that’s not protectable IP in and of itself CISC? In this economy?
|
|
#
?
Dec 25, 2023 02:49
|
|
|
Finally, a true successor to the Quadra 840AV
|
|
#
?
Dec 25, 2023 03:56
|
|
|
Altivec rides again
|
|
#
?
Dec 25, 2023 04:07
|
|
|
I do think that ColdFire is still being produced by NXP.
|
|
#
?
Dec 25, 2023 15:40
|
|
|
I’ve increasingly been seeing little ESP32 analogues built on RISC-V. is that because new-shiny targeted at compulsively acquisitive nerds, or does RISC-V offer some power/area advantage that’s worthwhile? the pricing is hard for me to compare because I can’t find any pair that are the same except core architecture
|
|
#
?
Dec 25, 2023 18:48
|
|
|
espressif have always been into weird architectures, the older ESPs were based on xtensa IP and then they moved onto RISC-V for newer parts
|
|
#
?
Dec 25, 2023 18:51
|
|
|
I thought I read at some point it was… questionably licensed Xtensa but on paper, RISC-V implementations should be “cheaper” with no ARM license costs. Now of course, ecosystem support etc etc on top of that… but I also have no idea how much a Cortex-M0/4 license actually costs.
|
|
#
?
Dec 25, 2023 18:59
|
|
|
ecosystem-wise even RISC-V is a step up for them, previously they were maintaining their own LLVM fork with xtensa support but their RISC-V stuff can use upstream
|
|
#
?
Dec 25, 2023 19:01
|
|
|
https://www.tomshardware.com/pc-com...pires-this-year The URL really says it all Edit: except that urls get shortened; uh, "Windows on Arm exclusivity may be a thing of the past soon — Arm CEO confirms Qualcomm's agreement with Microsoft expires this year" Also this at the end (can't find the Reuters article) quote:The exact date the exclusivity arrangement ends isn’t precise, but it will seemingly be gone by the start of 2025.
|
|
#
?
Jan 14, 2024 05:24
|
|
|
Hadlock posted:https://www.tomshardware.com/pc-com...pires-this-year It’s really hard to take the article seriously when the example they give of an NVIDIA SoC is a random server chip instead of the loving switch.
|
|
#
?
Jan 14, 2024 05:28
|
|
|
Reuters reported the end of of the exclusivity deal back in October, but I guess this is the first official comfirmation.
|
|
#
?
Jan 14, 2024 05:36
|
|
|
Yeah Tom's Hardware is not my first choice for balanced and well researched or fact checked journalism but the reference to Reuters (thanks for finding it) and an interview with AMD's CEO gives it some weight
|
|
#
?
Jan 14, 2024 05:48
|
|
|
I ran into a few articles on toms guides that were just straight up written with chatgpt a while back.
|
|
#
?
Jan 14, 2024 09:09
|
|
|
If anyone was going to tinker with that poo poo, it's gonna be a tech rag, yeah. Most tech writers seem to be transient 20-somethings who move of to other things so it's probably an improvement
|
|
#
?
Jan 14, 2024 09:30
|
|
|
|
| # ? Apr 29, 2024 02:02 |
|
|
What are some worthwhile sites for written articles these days?
|
|
#
?
Jan 14, 2024 10:51
|
|