Re: From sand to silicon chips, openly
On Fri, May 18, 2018 at 03:15:16AM -0400, grarpamp wrote: > > Where's the profit motive to develop things that can be sold on basis of > > value when the IP value is destroyed using an open architecture model? > > > > It's a lot of work to develop a CPU design and it costs a lot of money > > (well north of $50M). Who pays for that when the value is destroyed by > > opening up the design so anyone can copy it? > > > > Seems very naïve relative to how the semiconductor investment cycle works. There is at least some food for thought in comparison to the GNU project and the Linux kernel. 20 years ago it was of course “why would programmers give away their "intellectual property"?” which many at the time found a genuinely perplexing question - unfathomable if you will. Today, that story has just slightly/subtly, changed - Linux Everywhere and free/ libre/ open source software is basically a pre-requisite for any new venture or project. But why? - Why, as a programmer, would you give up your sovereignty/ right to make use of your own work in the future (either with another company, or in your own projects)? - Why would a company not take advantage of the $billions worth of free software "off the shelf" (and why would you spend millions reinventing all that NIH stuff)? - Why would you risk a proprietary startup when you know that the first guy to launch a FLOSS equivalent will in relatively short order (if you project/product is successful) overtake you in technical capability, on a relentless march? - If Microsoft couldn't successfully hold out against the FLOSS tide, what makes you think your puny thing will be able to? … and many similar besides. Create your world,
From sand to silicon chips, openly
> Where's the profit motive to develop things that can be sold on basis of > value when the IP value is destroyed using an open architecture model? > > It's a lot of work to develop a CPU design and it costs a lot of money (well > north of $50M). Who pays for that when the value is destroyed by opening up > the design so anyone can copy it? > > Seems very naïve relative to how the semiconductor investment cycle works. Asking on list would probably yield some of what you seek.
Re: From sand to silicon chips, openly
On Tue, May 15, 2018 at 12:39 AM, Steven Schearwrote: > A good example of why totally open chips are problematic in the commercial > world. > > Spectre/Meltdown Pits Transparency Against Liability: Which is More > Important to You? > https://www.bunniestudios.com/blog/?p=5127 > > As always, the devil is in the details. > > " You can’t have it both ways: the whole point of transparency is to enable > peer review, so you can find and fix bugs more quickly. But if every time a > bug is found, a manufacturer had to hand $50 to every user of their product > as a concession for the bug, they would quickly go out of business. This > partially answers the question why we don’t see open hardware much beyond > simple breakout boards and embedded controllers: it’s far too risky from a > liability standpoint to openly share the documentation for complex systems > under these circumstances. " As an incomplete snip from article, it would be bullshit on its own. At least for systems that start their life as open. Closed hardware / software generally asserts its fitness, at least in the corporate sales pitch, so when it fails you can sue the fuck out of them. For example, Intel is sued in court over Meltdown right now. Open fabs / hardware / software / dev pushes that entire analysis out to the user... they can inspect it, pay for analyst verification, read reviews, etc. In an open model, it becomes understood that all that is upon you, and the recourse is no longer suit, but filing bugs and commits and process change, and then the next release happens. That paradigm shift is the exact same in open fabs and hardware as it is in open software, even now in currencies and markets. You don't see it today because they want power, profit, control, and to them closed over the ignorant is the way to achieve that. After all, to date all the sheep have accepted that model of abuse. Openness and sharing is now proving in demand, profitable, and hopefully slowly taking over. Similar conclusions were in the article... "The Choice: Truthful Mistakes or Fake Perfection?" The offer is on the table. Even corporate users of HW know their fitness lawsuits etc will not always win and recover losses, so they'd also be insane not to offer it. > " However, even one of their most ardent open-source advocates pushed back > quite hard when I suggested they should share their pre-boot code. By > pre-boot code, I’m not talking about the little ROM blob that gets run after > reset to set up your peripherals so you can pull your bootloader from SD > card or SSD. That part was a no-brainer to share. I’m talking about the code > that gets run before the architecturally guaranteed “reset vector”. A number > of software developers (and alarmingly, some security experts) believe that > the life of a CPU begins at the reset vector. In fact, there’s often a > significant body of code that gets executed on a CPU to set things up to > meet the architectural guarantees of a hard reset – bringing all the > registers to their reset state, tuning clock generators, gating peripherals, > and so forth. Critically, chip makers heavily rely upon this pre-boot code > to also patch all kinds of embarrassing silicon bugs, and to enforce binning > rules." > > If, OTOH, there were ways to manufacture arbitrarily complex chips on the > desktop for reasonable costs and in reasonable time, and so eliminate the > commercial issues, this conundrum could vanish. > > > > On Wed, Mar 21, 2018 at 9:13 AM, Steven Schear > wrote: >> >> http://parallel.princeton.edu/openpiton/open_source_processors.php Nice list. Where's Intel and AMD and Qualcomm and ...
Re: From sand to silicon chips, openly
A good example of why totally open chips are problematic in the commercial world. Spectre/Meltdown Pits Transparency Against Liability: Which is More Important to You? https://www.bunniestudios.com/blog/?p=5127 As always, the devil is in the details. " You can’t have it both ways: the whole point of transparency is to enable peer review, so you can find and fix bugs more quickly. But if every time a bug is found, a manufacturer had to hand $50 to every user of their product as a concession for the bug, they would quickly go out of business. This partially answers the question why we don’t see open hardware much beyond simple breakout boards and embedded controllers: it’s far too risky from a liability standpoint to openly share the documentation for complex systems under these circumstances. " " However, even one of their most ardent open-source advocates pushed back quite hard when I suggested they should share their pre-boot code. By pre-boot code, I’m not talking about the little ROM blob that gets run after reset to set up your peripherals so you can pull your bootloader from SD card or SSD. That part was a no-brainer to share. I’m talking about the code that gets run before the architecturally guaranteed “reset vector”. A number of software developers (and alarmingly, some security experts) believe that the life of a CPU begins at the reset vector. In fact, there’s often a significant body of code that gets executed on a CPU to set things up to meet the architectural guarantees of a hard reset – bringing all the registers to their reset state, tuning clock generators, gating peripherals, and so forth. Critically, chip makers heavily rely upon this pre-boot code to also patch all kinds of embarrassing silicon bugs, and to enforce binning rules." If, OTOH, there were ways to manufacture arbitrarily complex chips on the desktop for reasonable costs and in reasonable time, and so eliminate the commercial issues, this conundrum could vanish. On Wed, Mar 21, 2018 at 9:13 AM, Steven Schearwrote: > http://parallel.princeton.edu/openpiton/open_source_processors.php >
Re: From sand to silicon chips, openly
On 3/28/2018 4:43 AM, juan wrote: On Sun, 25 Mar 2018 00:09:00 + Steven Schearwrote: Its one thing to have open CPUs but what about the rest of the hardware you need to build actual products? The best answers today are probably only be found in Shenzhen. > https://m.youtube.com/watch?v=SGJ5cZnoodY Some interesting, anti IP thoughts in there. ^ here's more stuff from bunnie huang more ed2k://|file|The%20Hardware%20Hacker_%20Adventures%20in%20Making%20and%20Breaking%20Hardware%20-%20Andrew%20_bunnie_%20Huang%20(No%20Starch%20Press;2017;9781593277581;eng).pdf|31906259|1ADD4CA296098F73CB070E31FA2FB863|/ A short time ago, China was well behind us in computer hardware, and was cloning and copying by rote. Lots of people predicted that as they caught up with us, their rate of progress would slow down, that they would always be one step behind. China is now nine months ahead of us in computer hardware, and the gap is growing. Notice that everyone can name the Soviet and US spy agencies, but no one can name the Chinese agencies, not even members of the Chinese spy agencies, which have far too many names, and no name at all. Some of the US aligned deep state agencies, the five eyes, have come to be entirely dominated by the US. The Canadian spy agencies appear to be completely dominated by the US agencies, though as US power declines, we could see something dramatic happen - or more likely something dramatic will silently happen, and it will be hidden from us, until we notice that US State Department organized protests, like the children's crusade against guns, are no longer echoed in Canada. The Australian deep state appears to be hedging its bets. From time to time it does things that make the Blue Empire very unhappy. It is currently in a state of grave conflict with the Chinese spy apparatus, but is keeping its options open for a re-alignment from the US to China, having both friendly and unfriendly contact with the quietly tolerated Chinese spy apparatus in Australia. Australia has a cadre of diplomats that are owned by the US blue state, plus some diplomats that owned by the Australian deep state, the Australian branch of the five eyes, resulting frequent conflict within Australia's diplomatic apparatus.
Re: From sand to silicon chips, openly
On Sun, 25 Mar 2018 00:09:00 + Steven Schearwrote: > Its one thing to have open CPUs but what about the rest of the > hardware you need to build actual products? The best answers today > are probably only be found in Shenzhen. > > https://m.youtube.com/watch?v=SGJ5cZnoodY Some interesting, anti IP thoughts in there. ^ here's more stuff from bunnie huang more ed2k://|file|The%20Hardware%20Hacker_%20Adventures%20in%20Making%20and%20Breaking%20Hardware%20-%20Andrew%20_bunnie_%20Huang%20(No%20Starch%20Press;2017;9781593277581;eng).pdf|31906259|1ADD4CA296098F73CB070E31FA2FB863|/ " every January, instead of going to the frenzied Consumer Electronics Show (CES) in Las Vegas, I rented a cheap apartment in Shenzhen and engaged in the “monastic study of manufacturing”; for the price of one night in Las Vegas, I lived in Shenzhen for a month. I deliberately picked neighborhoods with no English speakers and forced myself to learn the language and customs to survive. (Although I’m ethnically Chinese, my parents pri- oritized accent-free fluency in English over learning Chinese" > > > On Wed, Mar 21, 2018, 9:13 AM Steven Schear > wrote: > > > http://parallel.princeton.edu/openpiton/open_source_processors.php > >
Re: From sand to silicon chips, openly
Original message From: Steven Schear <schear.st...@gmail.com> Date: 3/24/18 5:09 PM (GMT-08:00) To: cypherpunks <cypherpunks@lists.cpunks.org> Subject: Re: From sand to silicon chips, openly Its one thing to have open CPUs but what about the rest of the hardware you need to build actual products? The best answers today are probably only be found in Shenzhen. https://m.youtube.com/watch?v=SGJ5cZnoodY Welp thats all done then, considering China is smacking Trump's ass about tariffs. Just today one of the bay area papers was pointing out the connection between their ability to tariff amost everything related to the electromechanical end of the computer industry and Google/Twiitter/Facebook engineers sleeping in a refrigerator box by the river. Rr On Wed, Mar 21, 2018, 9:13 AM Steven Schear <schear.st...@gmail.com> wrote: http://parallel.princeton.edu/openpiton/open_source_processors.php
Re: From sand to silicon chips, openly
Its one thing to have open CPUs but what about the rest of the hardware you need to build actual products? The best answers today are probably only be found in Shenzhen. https://m.youtube.com/watch?v=SGJ5cZnoodY On Wed, Mar 21, 2018, 9:13 AM Steven Schearwrote: > http://parallel.princeton.edu/openpiton/open_source_processors.php >
Re: From sand to silicon chips, openly
On Wed, Mar 21, 2018 at 09:13:02AM -0700, Steven Schear wrote: > http://parallel.princeton.edu/openpiton/open_source_processors.php Looks like some nice architecture stuff being explored here, reducing core-count coherence overhead ("up to 64" core coherence domains are optimized) and reducing power usage with "drafting mode" which is essentially SIMD applied at a higher level ('cohering' many threads - e.g. uploading and analysing/scaling images for many web page end users - thread instructions are aligned to execute simultaneously (where possible) improving throughput to energy by up to 20 percent, 8.57 percent on average). So they're targetting the right metrics for a longer term traction - "throughput divided by energy" which is what datacenters must optimize if they want to increase profitability - a little different to minimizing latency, and minimizing power usage per data processed/ stuff done, required for battery powered end user devices - although I think it's not yet clear that these two outcomes are not ultimately one and the same result (with a laptop, hike the CPU frequency when some heavy calculations need to be done, to minimize the time required to do them (user's want low latency) and this thereby also minimizing overall power usage). All roads lead to Rome.
Re: From sand to silicon chips, openly
Original message From: Steven Schear <schear.st...@gmail.com> Date: 3/21/18 9:13 AM (GMT-08:00) To: cypherpunks <cypherpunks@lists.cpunks.org> Subject: From sand to silicon chips, openly http://parallel.princeton.edu/openpiton/open_source_processors.php Say buhbye... Peak Sand https://www.google.com/search?q=peak+sand Your grandchildren will be using an abacus. Rr
From sand to silicon chips, openly
http://parallel.princeton.edu/openpiton/open_source_processors.php