|
Yah that password checks out, I would actually get intel alpha/beta firmware zipped up with that password or similar.
|
#
?
Aug 6, 2020 20:22
|
|
|
|
| # ? Jun 3, 2024 00:12 |
|
|
Usually everyone should have it drilled in to them to protect this poo poo but all plans fall apart when the secure file site takes a poo poo or people can't figure it out and things end up on an open network share
|
|
#
?
Aug 6, 2020 20:27
|
|
|
Yup, it’s kind of archaic to just be emailing large files around pw protected zip or no. Have some secure as possible cloud share thing you can grant and revoke access to at the very least! Even internally we have so much trouble sharing large files with the way IT has everything locked down there is a tough balance between security and usability.
|
|
#
?
Aug 6, 2020 20:37
|
|
|
priznat posted:Yup, it’s kind of archaic to just be emailing large files around pw protected zip or no. Have some secure as possible cloud share thing you can grant and revoke access to at the very least! My previous group had a secure file site with permission management and all that. But it requires Java. IT came in and removed Java from everyone's laptops one night and the only way we could access our docs was by using department not IT administered lab systems
|
|
#
?
Aug 6, 2020 20:40
|
|
|
WhyteRyce posted:My previous group had a secure file site with permission management and all that. But it requires Java. IT came in and removed Java from everyone's laptops one night and the only way we could access our docs was by using department not IT administered lab systems We had to set up our own server in the lab for storing traces etc so other groups could access them because the IT provided storage requires an arduous process that takes 2-3 days to add new people to it and if they don’t log in for 30 days their access is removed. So every project we would have a bunch of people clamouring to get our traces but no way to provide them. It worked well using our own server until IT hosed with the domain accesses and now if you are coming in via certain VPNs you can’t access it. But this is getting a bit off topic should go in an IT bitching thread 
|
|
#
?
Aug 6, 2020 20:52
|
|
|
Can someone smarter than me describe what's in that leak and why it's a big deal
|
|
#
?
Aug 6, 2020 20:59
|
|
|
gradenko_2000 posted:Can someone smarter than me describe what's in that leak and why it's a big deal It just dropped, so there's a lot that hasn't been analyzed yet. But part of the concern is that it may expose information that could be used to build malware against chips whose BIOS code is included.
|
|
#
?
Aug 6, 2020 21:10
|
|
|
gradenko_2000 posted:Can someone smarter than me describe what's in that leak and why it's a big deal A shitload of proprietary Intel motherboard/ME source, binaries, internal development and debugging tools, documentation, roadmaps, and even chip designs.
|
|
#
?
Aug 6, 2020 21:11
|
|
|
Dr. Fishopolis posted:Okay sure but literally no person, not a single soul is buying a mac pro to train 1tb+ datasets. Your argument makes sense from a processor market segmentation standpoint but Apple is selling something that's ostensibly a media production workstation. There shouldn't even be xeons in there to begin with. This audio freelancer dude would disagree with you on that, this is a long video series but iirc he ends up being able to load up pretty much all cores of his new mac pro audio workstation setup. Also iirc he ends up needing a substantial amount of RAM. Not 1TB+, but still more than you might expect. Media workstations can need lots of storage, memory, and compute. https://www.youtube.com/playlist?list=PLi2i2YhL6okTb7-pQbRk9rSWrw0JknLxT
|
|
#
?
Aug 6, 2020 21:19
|
|
|
There are apparently some fun things in this leak:pre:- Intel ME Bringup guides + (flash) tooling + samples for various platforms - Kabylake (Purley Platform) BIOS Reference Code and Sample Code + Initialization code (some of it as exported git repos with full history) - Intel CEFDK (Consumer Electronics Firmware Development Kit (Bootloader stuff)) SOURCES - Silicon / FSP source code packages for various platforms - Various Intel Development and Debugging Tools - Simics Simulation for Rocket Lake S and potentially other platforms - Various roadmaps and other documents - Binaries for Camera drivers Intel made for SpaceX - Schematics, Docs, Tools + Firmware for the unreleased Tiger Lake platform - (very horrible) Kabylake FDK training videos - Intel Trace Hub + decoder files for various Intel ME versions - Elkhart Lake Silicon Reference and Platform Sample Code - Some Verilog stuff for various Xeon Platforms, unsure what it is exactly. - Debug BIOS/TXE builds for various Platforms - Bootguard SDK (encrypted zip) - Intel Snowridge / Snowfish Process Simulator ADK - Various schematics - Intel Marketing Material Templates (InDesign) - Lots of other things
|
|
#
?
Aug 6, 2020 21:31
|
|
|
Maybe hackers could do us a favour and improve the ITP suite
|
|
#
?
Aug 6, 2020 21:33
|
|
|
priznat posted:Maybe hackers could do us a favour and improve the ITP suite you've triggered me
|
|
#
?
Aug 6, 2020 21:42
|
|
|
WhyteRyce posted:you've triggered me   Schedule estimates: - ITP test setup 2 days - Licensing issues 3 weeks - ITP test run 2 days Is someone good with scheduling please help my schedule is dying
|
|
#
?
Aug 6, 2020 21:44
|
|
|
That dump looks a lot like what an integrator would pull from a portal; I would get Nehalem CRB, MRC and other stuff from a portal like that. If they canary trapped any of that they might find what dumbshit decided to upload that. Didn’t think there’s any actual RTL in that dump but haven’t looked myself (I see it’s called out but if that’s real real stuff, who the gently caress had access to THIS combination of poo poo?). The real take away of course is that Intel has a code name for EVERYTHING. Even the dumbest, simplest little eval boards would show up with “Turtle Cove 7” and corny artwork on it. E: lol did they leak a Quartus manual
|
|
#
?
Aug 6, 2020 21:52
|
|
|
movax posted:That dump looks a lot like what an integrator would pull from a portal; I would get Nehalem CRB, MRC and other stuff from a portal like that. If they canary trapped any of that they might find what dumbshit decided to upload that. Yep that's what Intel has said. quote:A spokesperson for Intel said the information was likely taken from its Resource and Design Center, which is a private library of resources for computer manufacturers and the like to build systems using Intel's silicon. Access to this center is not open to the public as its content is intended to be used, for instance, to craft firmware and design motherboards compatible with Chipzilla's microprocessors. quote:The purported source of the leaked documents has explained how they found the material: apparently by scanning the internet with Nmap and finding an unsecured Akamai CDN server hosting Chipzilla's files.
|
|
#
?
Aug 6, 2020 22:00
|
|
|
movax posted:That dump looks a lot like what an integrator would pull from a portal; I would get Nehalem CRB, MRC and other stuff from a portal like that. If they canary trapped any of that they might find what dumbshit decided to upload that. it makes far more sense that someone compromised a dumbass integrator than actually breached internal intel accounts
|
|
#
?
Aug 6, 2020 22:22
|
|
|
BobHoward posted:This audio freelancer dude would disagree with you on that, this is a long video series but iirc he ends up being able to load up pretty much all cores of his new mac pro audio workstation setup. Okay, that's a fair enough use case, if you're live tracking to a huge VST orchestra and you don't want to bounce down first I can see it. That's one place where macos has a huge software and stability advantage, a mac pro or two in a big touring rig makes a lot of sense. e: threadripper would still be a better price/performance choice but no pro is gonna rely on an amd hackintosh. Dr. Fishopolis fucked around with this message at 22:27 on Aug 6, 2020 |
|
#
?
Aug 6, 2020 22:24
|
|
|
D. Ebdrup posted:There are apparently some fun things in this leak: i literally can't wait for these to show up on youtube
|
|
#
?
Aug 6, 2020 22:29
|
|
|
In case this is not obvious: the passworded zips sharing the same intel123 password are simply to circumvent their anti-virus/mail server checker. I've been given pre-release binaries with an Intel person I was collaborating with and it was all zipped in that same way.
Beef fucked around with this message at 22:45 on Aug 6, 2020 |
|
#
?
Aug 6, 2020 22:38
|
|
|
I just want to see the (very horrible) Kabylake FDK training videos
|
|
#
?
Aug 6, 2020 22:46
|
|
|
The PDK documentation is hilariously minimalist. It has a full brochure going over the safe way to lift a large heavy server but nothing when it comes to things like accessing bios, updating firmware, etc.
|
|
#
?
Aug 6, 2020 22:53
|
|
|
Dr. Fishopolis posted:i literally can't wait for these to show up on youtube likewise the amount of passworded archives is overrated as well, for the 100+ in there 9 have passwords. they seem to love bundling software with their decryption keys though. nothing too spectacular in the dump outside of mild comedy and nda'ed tools for testing pre/post silicon. i'm not a low levels engineer though so maybe someone in that security field will find a relevant gem in the pile, the ME collection isn't anything special outside of integrators
|
|
#
?
Aug 7, 2020 01:24
|
|
|
The dump with git history seems fun though. Don't know how many times I used stupid commit messages on internal tools because I assumed no one important actually reads the logs
WhyteRyce fucked around with this message at 01:50 on Aug 7, 2020 |
|
#
?
Aug 7, 2020 01:43
|
|
|
I built a fully functioning hackintosh on a x570 board and it actually runs great for most things. However, there is a big butt, and that comes in the form of pro software. The current version of Adobe Lightroom flat out doesn't work. You can delete some files in Photoshop to make it work, but other apps in the CC are also suspect and unreliable. I've heard of issues with FCP as well. These issues make sense when you can target and optimize only for Intel because Apple does not make AMD Macs. In daily use, I don't have any issues and my MacFaux7,1 is loving great with 12 cores, 24 threads, and 32gb of ram. But those techtubers leave out poo poo like app compatibility in those bake offs for a reason.
|
|
#
?
Aug 7, 2020 02:53
|
|
|
I wish we got the source to the IME
|
|
#
?
Aug 7, 2020 03:30
|
|
|
priznat posted:Maybe hackers could do us a favour and improve the ITP suite Lol
|
|
#
?
Aug 7, 2020 05:56
|
|
|
https://www.pcmag.com/news/intel-is-releasing-avengers-edition-10th-gen-processors so it turns out the "KA" suffixed processors that we saw some weeks ago stands for "Avengers" - they're exactly the same as the base models, but they're going to be packaged in Marvel: Avengers-themed boxes
|
|
#
?
Aug 7, 2020 13:31
|
|
|
gradenko_2000 posted:https://www.pcmag.com/news/intel-is-releasing-avengers-edition-10th-gen-processors
|
|
#
?
Aug 7, 2020 13:47
|
|
|
ConanTheLibrarian posted:Appropriate that Intel's attempt to tie in to an Avengers movie turns up over a year late. 
|
|
#
?
Aug 7, 2020 14:04
|
|
|
gradenko_2000 posted:https://www.pcmag.com/news/intel-is-releasing-avengers-edition-10th-gen-processors  ConanTheLibrarian posted:Appropriate that Intel's attempt to tie in to an Avengers movie turns up over a year late. 
|
|
#
?
Aug 7, 2020 21:42
|
|
|
I think it’s a tie in with the game that’s about to come out, but the joke is better please ignore!!
|
|
#
?
Aug 7, 2020 21:52
|
|
|
Do we expect to see Tiger Lake / Rocket Lake have any significant improvements for desktop users? The only thing that looks appealing is PCI Express 4.0 but not much else at all.
|
|
#
?
Aug 9, 2020 00:54
|
|
|
The first new core arch/IPC gain since 2015.
|
|
#
?
Aug 9, 2020 01:16
|
|
|
Yeah who knows what that actually means, could be irrelevant, could be big. Just because Intel has completely poo poo the bed on process tech doesn't mean that they've forgotten how to do architectural improvements. But on the other hand, Intel company culture in general seems pretty rotten at this point so who knows.
|
|
#
?
Aug 9, 2020 01:17
|
|
|
Gimmie Sapphire Rapids DDR5/PCIe5 
|
|
#
?
Aug 9, 2020 01:20
|
|
|
priznat posted:Gimmie Sapphire Rapids DDR5/PCIe5 with support for Optane Persisntent Memory so I don't even need an nvme SSD
|
|
#
?
Aug 9, 2020 04:27
|
|
|
Cygni posted:The first new core arch/IPC gain since 2015. I might be out of the loop but has nothing changed since Skylake? That doesn't make sense...
|
|
#
?
Aug 9, 2020 05:14
|
|
|
Gabriel S. posted:I might be out of the loop but has nothing changed since Skylake? That doesn't make sense... I think that covers most of the major changes since skylake. Nothing about the architecture has really changed. IPC is identical. The big change is you don't have to delid & overclock to get near peak performance anymore. It's just there out of the box. People who bought 8700ks near release made out like the sandy & ivy bridge bandits unless they're big into streaming or productivity tasks that benefit from tons more cores. Ice Lake and the 10nm family CPUs have some really nice ipc gains, but their clock speed is pretty low so they underperform vs the sky lake derivatives on desktop (& aren't available anyway). For laptops, it's a complicated minefield that I'm not stepping into in this quick reply. Khorne fucked around with this message at 05:27 on Aug 9, 2020 |
|
#
?
Aug 9, 2020 05:23
|
|
|
Intel has long operated on a tick-tock model, where they introduce a new architecture on an existing process, then port that architecture to a new process, and repeat. This first started to slip at 22nm, when 14nm wasn't ready to go on time and Intel did the Haswell refresh, i.e. replacing the 4670k with the 4690k and the 4770k with the 4790k. Same architecture, same process, just a frequency bump. At 14nm, this completely fell apart. Broadwell was Haswell ported to 14nm, and then Skylake a year later was a new architecture on that 14nm process. The next step would be the port Skylake to 10nm and then release a new arch on 10nm. Intel spent years believing that 10nm was just around the corner, so all their major architectural development work was done planning for 10nm. 10nm came years late and it sucks in basically every way. So for years Intel has done nothing but re-release the same architecture on the same process with optimizations and some new features (mainly increased core counts). Something like a year ago Intel came to accept that 10nm was a failure and would probably never be good, so while waiting for 7nm they decided to take architectural improvements that they'd designed for 10nm and backport them to new CPUs designed for 14nm. Rocket Lake will be the first series of processors based on this new architecture.
|
|
#
?
Aug 9, 2020 05:33
|
|
|
|
| # ? Jun 3, 2024 00:12 |
|
|
It looks like though that it's just that we'll see Tiger Lake (Mobile) and Rocket Lake (Desktop) not until later in 2021. Or maybe just a paper launch.
|
|
#
?
Aug 9, 2020 05:43
|
|