SoylentNews is people
SoylentNews
from the watch-for-bugs-at-the-chip-buffet dept.
High Performance Computing (HPC) Chips – A Veritable Smorgasbord?
No this isn't about the song from Charlotte's Web or the Scandinavian predilection for open sandwiches; it's about the apparent newfound choice in the HPC CPU market.
For the first time since AMD's ill-fated launch of Bulldozer the answer to the question, 'Which CPU will be in my next HPC system?' doesn't have to be 'Whichever variety of Intel Xeon E5 they are selling when we procure'.
In fact, it's not just in the x86 market where there is now a genuine choice. Soon we will have at least two credible ARM v8 ISA CPUs (from Cavium and Qualcomm respectively) and IBM have gone all in on the Power architecture (having at one point in the last ten years had four competing HPC CPU lines – x86, Blue Gene, Power and Cell).
In fact, it may even be Intel that is left wondering which horse to back in the HPC CPU race with both Xeon lines looking insufficiently differentiated going forward. A symptom of this dilemma is the recent restructuring of the Xeon line along with associated pricing and feature segmentation.
(Score: 0) by Anonymous Coward on Sunday October 15 2017, @07:17PM (6 children)
IBM's superior POWER architecture hangs in there, and then suddenly, a RISC-V group begins producing chips that are interesting. IBM says "Aw, fuck it!", and puts all of its POWER resources behind the RISC-V project; this encourages the likes of Google and Facebook to go all in on the RISC-V system, and then the whole world benefits as this open-hardware technology trickles down to consumer products.
(Score: 0) by Anonymous Coward on Sunday October 15 2017, @07:49PM (3 children)
RISC-V will end up like MIPS, rather than like ARM, due to too much flexibility allowed in the per-chip ABIs.
The J-series of SuperH clone processors seems a more likely bet in my book, but they need to tape out a 'Pi alternative' chip in the next two years or less and either 64 bit addressing extensions or an SH5 equivalent (since AFAIK SH5 is still under patents and gcc just removed support..) ASAP for any of it to still be relevant. Same problem RISC-V has today: There are no serious plans for making a RISC-V desktop ecosystem, least of all a 'true' multi-source one with intercompatible components and royalty free cross licensing agreements.
(Score: 3, Informative) by RamiK on Sunday October 15 2017, @10:34PM
No. RISC-V standardized the extension mechanism to prevent the per-chip ABIs fragmentation that crippled MIPS (off the v2 draft ISA manual):
Why? Once the patent is up you could see just as much fragmentation. Only, unlike RISC-V, there's no mechanism to deal with someone adding extra instructions.
SiFive Freedom U500 can be evaluated on a $3.5k FPGA kit and should fit the RasPi3\ODROID-C2 space once produced on both cost and performance.
The x86 desktop market is a high-risk, low-reward venture. You're restricted to 4 instructions wide out-of-order superscalar pointer addressing designs since there's too much code to rewrite otherwise. Closest I've seen anyone coming near it was ARM's recent A75 that's targeting smartphones.
compiling...
(Score: 2) by driverless on Monday October 16 2017, @10:13AM (1 child)
Nah, RISV-V will end up like the Motorola 88000, after some initial noise it'll fade slowly into oblivion, accompanied by the sound of no-one caring.
And I don't mean that as a snarky comment, it's an also-ran with no support infrastructure outside of one or two obscure vendors, why would a device designer go with one of these? You're betting your product, and possibly your company, on a barely-there product that may not exist next month. There's no way I could convince any of my mgt to go with something like this for anything outside of a hobby project done on my own time.
(Score: 0) by Anonymous Coward on Monday October 16 2017, @04:22PM
RISC-V is something that will grow from niche to mainstream.
You fuckers are always trying to imagine lucrative ventures, but that's not where this will start. As always, it will start in hackers' proverbial garages.
(Score: 0) by Anonymous Coward on Sunday October 15 2017, @08:36PM (1 child)
Nah, they'll send in the Nazgul making claims of patent infringement and other forms of Imaginary Property theft. ARM And intel will shake their hands and do some more cross-licensing with IBM having one less competitor and the little people lose.
(Score: 2) by bob_super on Tuesday October 17 2017, @04:24PM
That's because most people are oblivious as they stare at Kim K's ass on their ARM or x86 machines.
Now, if you paid enough money for Kim K to release her latest nudies in a way that only decodes on a specific platform, you might see either an uptick of interest, or people starting ignoring her for free porn elsewhere.
Both are very positive outcomes, where's the GoFundMe page?