Extremely disappointing.
Also the CES didn't say anything about the next Threadrippers. :(
They did tease the next Zen architecture but not much details sadly.
But having Microsoft-designed hardware as a backdoor in my system? Absolutely not.
Also the CES didn't say anything about the next Threadrippers. :(
They did tease the next Zen architecture but not much details sadly.
But having Microsoft-designed hardware as a backdoor in my system? Absolutely not.
Sigh, less and less choice every day.
EDIT: What about this potential leak? https://www.digitaltrends.com/computing/amd-to-release-ryzen...
Other than that, performance is plenty on a 5950x for all my imaginable use cases.
Reliability
ECC Memory
To protect against data corruption
Error Correction Code (ECC) memory corrects errors in your data
as it passes in and out of memory to ensure reliability for
critical applications.
I went with an Asus instead. The Asus manual says "ECC memory support varies by CPU" which is questionable for the 5700G APU I went with, and I didn't feel like digging to find out if harder to obtain ECC UDIMMs would be beneficial. (This was for a router / personal server. My workstation is ECC on a libreboot/KGPE)
Is it all-AMD? If it is, could you link to your setup? I am looking to build an all-AMD workstation this year and I need all the PCIe lanes that I can get.
https://libreboot.org/docs/hardware/kgpe-d16.html
https://www.coreboot.org/Board:asus/kgpe-d16
It's got 5 usable PCIe slots (including the physically-flipped "PIKE" one), I think a total of 40 PCIe lanes from both sockets that I believe you could theoretically split out with the maximum amount of bifurcation, since the BIOS is Free Software.
My new Ryzen build is for a personal server that resides at a lower trust level, and targets around 30W draw. B550 motherboards generally have around 28 PCIe lanes going to expansion slots (20 from the proc, 8 from the chipset). The best you can do for number of slots is to bifurcate the main "graphics" x16 into 3 x4 slots (maybe 4 with a non-APU) using a "VROC" card, but the BIOS has to support that (Gigabyte and Asus seem to generally, but check manual). I believe X570 motherboards have a few more PCIe lanes coming from the chipset.
If you really need more PCIe lanes, I hear Threadripper/EPYC is the way to go. But I don't have any personal experience. If you just need more PCIe slots, you can find PCIe x1 -> four slot switch-based splitters inexpensively on ebay/aliexpress.
> Last time I stuck a Kill-a-watt on it, I think it draws around 140W at idle
Another tangent (sorry!): do you know of good power meters that don't have to sit right at the outlet? I really need several but I want them to have extensions because I want to glue them on my wall and monitor them in real time, not having to stick my ass in the air while trying to crawl under my desk (where all the outlets are), just so I can see the measurements.
[0] it wants to connect to "cloud" as well, but works fine without giving it Internet access.
EDIT: miscalculates as rarely as a computer normally does
Thread Threadripper uses a massive amount of high quality silicon, Epyc has much better margins.
Supporting HEDT is living hell for Intel/AMD. You're often using all kinds of consumer hardware with otherwise enterprise grade gear. Addressing every little PCIe device issue and catering to a very very small market has very little ROI.
For something like Intel's HEDT then yes, you're absolutely correct. That involves HEDT-specific binning and it eats into Xeon-W profits very directly. But for Threadripper it's not nearly as clear cut. In fact it seems more like it's silicon that failed to validate for Epyc since it's half the IO die of Epyc (for the non-Pro Threadripper anyway).
Otherwise you're talking about AMD just taking Ryzen-quality chiplets, putting more of them on a substrate, and selling them at a huge markup.
Take the Threadripper 3970X as just a simple example here. It's 4x 8c zen2 chiplets that can hit a peak of 4.5ghz at 280W TDP. Meanwhile the Ryzen 3950x is a 2x 8c zen2 chiplet that hits a peak of 4.7ghz at 105W TDP. So 3970x is 2x the silicon of the 3950x, but lower quality silicon, and AMD charged more than 2x for the 3970X.
That's the brilliance of chiplets and Threadripper. It's Ryzen-class chiplets with half of an Epyc IO die (the IO die being 12nm GloFlo means it's not really competing for high-end prices anyway). It's binning they're already doing for they're primary product lines, not a specialized HEDT-specific process.
So unless AMD just doesn't want a piece of that pie anymore despite only just entering the market, there's absolutely more Threadrippers coming. No doubt about that at all. AMD isn't going to leave the Xeon-W just entirely unchallenged when AMD is otherwise firing on all cylinders.
What they probably won't do is continue to have different sockets for the "Pro" and "non-Pro" threadrippers, though. But that'd be different from them just not having anymore Threadrippers at all. It'll probably just end up in the same category as Ryzen & Ryzen Pro where it's the same socket, just one is sold to consumers and one is sold to OEMs.
Now, a Zen 4 Threadripper is extremely compelling because of DDR5 and PCIe 5.0.
I highly doubt we'll see a vcache version of TR/TR Pro prior to the jump to Zen 4, especially since so far AMD has only officially announced a 3d vcache 5800X which is a single-chiplet product ( https://www.anandtech.com/show/17152/amd-cpus-in-2022-ces )
I do wonder though, what improvements will they bring to the next Threadrippers? There are some rumours about the start of March 2022, I am very curious what can they unveil. Any theories?
And I am not sure Zen 4 Threadrippers would even be available this year but who knows. Plenty of rumours. I better start saving up money because I want to invest in a workstation that will last me at least 10 years (minus SSD wear). And Zen 4 / PCIe v5 / DDR-5 look preeeeeeeetty good.
I agree that a March launch would mean it would be on Zen 3.
I'd guess they planned on TR / TR Pro being a quicker followup after Epyc 7003 launched, but then chip capacity just wasn't there to justify it. And now it probably is, especially as their new stuff is on a different node.
The way I see it, Zen 2 Threadripper is still a good enough product for its goals of a shit load of cores. Zen 3 didn't increase core counts, but Zen 4 does. My suspicion is that AMD probably overproduced Zen 2 TR and is letting its supply run down until it can launch another category redefining product built on Zen 4. Plus, it would look even better in their comparison charts.
But AMD is going to want to establish consistency with OEMs still, so Intel not having anything competitive right now is merely a reduction in pressure on AMD, not something that eliminates it entirely. A Zen 3 Threadripper Pro as a result still seems more likely than not, even if consumer Threadripper doesn't get a Zen 3 treatment. I'm sure AMD would also prefer to simplify the chiplets its fabricating as well. Why keep making Zen 2 chiptlets for TR Pro specifically and nothing else?
edit: So, I'm sure there's some other older thoughts/info presented on his podcasts, but it's the latest thing I heard from him talking about the subject:
https://www.youtube.com/watch?v=b8gVyjvSZkg&t=5713s (12 minute section )
I think there is as of yet insuficient data to support that claim. A security processor that guarantees safety for critical data is a good piece of hardware for those that require such functionality, as long as it's optional, not-backrooted, not controlled by a remote corporation, has an open API and interactions with the system can be audited:
Windows devices with Pluton will use the Pluton security processor to protect credentials, user identities, encryption keys, and personal data. None of this information can be removed from Pluton even if an attacker has installed malware or has complete physical possession of the PC.
I don't disagree per se but:
1. I don't give corporations the positive benefit of the doubt. They have done way too many attacks on privacy and I think it'd be naive to think they won't get tempted to have a below-the-kernel backdoor access to everyone's systems. They absolutely would, given the chance. Let's not give them the chance is what I am advocating for.
2. It's not about the advertised good use of the tech. We should first and foremost look for the abuse potential. If it's there then the tech should be dropped. Failing that (because it's too idealistic, I realize that) then we should make double and triple sure our routers and switches are secure and will stop suspicious traffic (if that's even possible for a home user...). Or have systems in place that don't let in every single firmware update.
I am not the first to point out that malware writers are a big beneficiary of such features -- it's a great way to avoid inspection/detection of their code.
"as long as it's optional"
DRM systems ultimately become de facto mandatory, even if they remain optional in theory.
"not controlled by a remote corporation"
Hardware DRM systems ("trusted computing") are inherently controlled by remote corporations; at the end of the day someone has to certify hardware-bound keys, and someone has to revoke leaked keys. Companies routinely try to obscure this crucial detail but if you dig deeply enough in the documentation you will inevitably find it.
And that is not bad technology in itself to have, the question is who gets to sign, and it's a political one. There exist in principle an acceptable threshold that various small OS vendors and Linux/*BSD distributions can pass and virus authors cannot, while allowing for some sort of hardware three finger salute that enables custom self signed roots for organizations and developers that can handle them safely, i.e. a special use case not required by the majority of the population who simply trusts a vendor.
I don't claim the Microsoft tech allows these (most likely not), but I believe these are the correct demands and criticism we should make, approach it as a political issue.
Fighting ideologically against a technology and ignoring the larger political objectives is foolhardy. If the technology delivers value, some vendor will bundle it with pretty pink buttons and corner the market, forcing you to use it too because every body else does and you have no other option. It's how we ended up with solid blocks of DRM from Apple in the hands of billions of consumers that won't even allow you to run your own choice of software, let alone alter the operating system.
I just removed a crypto-miner malware from an IT-illiterate friend's windows computer. There are great many people in this category. If we can't fix their computers by flipping a TPM switch that ensures some level of platform integrity, they will simply go out and buy an Apple device, "because it works better". That's their subjective view, good luck teaching them to value software freedom and learn good security practices.
The technology does not deliver value for users. It has always been intended to benefit Microsoft and their media partners, with a bit of window-dressing meant to trick users into believing that they somehow gain from it. This is possible only because the current market for personal computers has almost no meaningful competition.
"I just removed a crypto-miner malware from an IT-illiterate friend's windows computer"
...and malware authors will use DRM systems like this to make it harder to detect their malware. Instead of, "Hm, CRYPTOMINER.EXE is spinning the CPU" it will be "Something seems to be spinning the CPU but the platform DRM is preventing me from figuring out what is happening."
"some level of platform integrity"
Except that this system does not ensure platform integrity. Sure, firmware and bootloader signing can protect against malware or at least give users the ability to reset their system to a good state, but we are talking about a DRM system and that is a very different story.
Of course, Microsoft's track record on bootloader security is mixed. They have been willing to allow major Linux distros to get a signed "shim" that can be used to bootstrap grub, but they also made a deliberate and arbitrary decision to forbid vendors of ARM systems from allowing users to disable secure boot. The result is that users who want to run their own bootloader, with whatever risk that entails, have less choice in hardware and are forced to spend more. So while UEFI was generally a win for end user security, it came with a strategic effort by Microsoft to exert greater control over user devices -- something which only benefits Microsoft and which was done only as part of their long-term effort to sell DRM to media companies.
The end of all this will be a world where everything looks like the "mobile" ecosystem or video game consoles -- users will not be allowed to run any software that Microsoft did not approve of unless they pay 4-5x more for a computer that has fewer restrictions. Sure, it will make life harder for malware writers -- assuming the approved software does not have tons of exploits -- but it will also mean that Microsoft's interests never get challenged. The only reason anyone will be allowed to run Libreoffice will be competition authorities, and in all likelihood software will be made available based on a user's region (so EU users get to run Libreoffice, but not US users). It will be a net negative for users and for the next generation of developers, who, rather than learning by staying up late with their own computers, will at best only be able to program on their school's computers and only if they are in a wealthy enough school district.
Spoiler alert, if you have any Intel CPU released after 2006-2008, you have one.
In the past few years AMD has started including a BIOS option to disable it. However, I have never seen a convincing explanation of how exactly that option works. The only thing I know is that Linux complains about it at boot on my B450M (from 2019):
Aug 30 23:52:07 kobold kernel: [ 4.811829] ccp 0000:07:00.1: ccp: unable to access the device: you might be running a broken BIOS.
Aug 30 23:52:07 kobold kernel: [ 4.811831] ccp 0000:07:00.1: psp: unable to access the device: you might be running a broken BIOS.
I entertained the idea of me_cleaner many times but yeah, I really can't afford to lose any of my machines and I can't trust almost anyone to do it properly except me -- and I can't do it.
Can you clarify on AMD's PSP? Did you start getting the message after you (supposedly) deactivated it in BIOS? Or is the message only visible when it's active? Or is it always visible?
You might consider picking up a cheap used motherboard from ebay or the like to experiment with, if you're really interested.
> Did you start getting the message after you (supposedly) deactivated it in BIOS?
Yes.