|
.
sincx fucked around with this message at 05:55 on Mar 23, 2021 |
#
?
May 17, 2019 05:57
|
|
|
|
| # ? May 20, 2024 04:48 |
|
|
I felt kind of paranoid buying my wife an 8/8 a few months back, but yeahhhh not really anymore  that is still a decent amount of cores.Why not just bring back two CPU motherboards, then they could sell twice as many CPS as well sadus fucked around with this message at 06:06 on May 17, 2019 |
|
#
?
May 17, 2019 06:03
|
|
|
It might be the case that the majority of users aren't realistically under threat but there's still uncertainty whether Intel or OS providers (Apple, Google, Microsoft) will force you to take a performance hit to cover their bases. ChromeOS disabling HT by default is a minor example but is Intel going to keep separate branches of its software/firmware updates moving forward for people that don't want to take any performance hit? They've already specifically stated that their soft mitigation causes up to an 8% performance hit.
|
|
#
?
May 17, 2019 06:51
|
|
|
It's only a matter of time, the writing is on the wall. We're going to keep having more and more of these exploits coming out. I'm sure Apple will be next to disable HT by default and then Microsoft will follow suit just to keep from being seen as the (even more) insecure OS.
|
|
#
?
May 17, 2019 15:40
|
|
|
sincx posted:Solution: buy AMD Except, as has already been noted, consumers aren't really much at risk, and the mitigations only minorly impact performance. For big data, where disabling HT might actually be painful, the cost of throwing out and replacing all your poo poo with enough AMD processors to get back to the same level of performance you had with Intel would be hilariously more expensive than just disabling HT / patching and buying a couple extra Intel boxes to make up the difference. Zen2 might change the calculus for people spec'ing out entirely new installs, but we'll see what they do with pricing, especially on the higher end "you're not so much buying chips as you are buying entire systems" type deals. I also am real curious to see if AMD ends up being vulnerable to similar prediction attacks at some point. Yes, yes, I know they're not vulnerable to these specific attacks, but that may very well be simply because the researchers were looking for ways to attack Intel chips, and the specific approaches taken so far don't map 1:1 onto AMD topography. But if AMD actually gets market share back, there may be more interest in spending time and effort targeting AMD chips to see what sort of interesting things they can wring out of them. Kinda like how Apple OS isn't magically immune to attacks and malware, but more that few people bother making malware for it because it's only something like 10% of the market. Or maybe AMD takes the lessons from these attacks and bakes protections into future chips before anyone gets around to poking at them much. We shall see.
|
|
#
?
May 17, 2019 16:23
|
|
|
shrike82 posted:It might be the case that the majority of users aren't realistically under threat but there's still uncertainty whether Intel or OS providers (Apple, Google, Microsoft) will force you to take a performance hit to cover their bases. ChromeOS disabling HT by default is a minor example but is Intel going to keep separate branches of its software/firmware updates moving forward for people that don't want to take any performance hit? They've already specifically stated that their soft mitigation causes up to an 8% performance hit. If you don't want the performance hit from the soft mitigation, you don't patch. That's it. You're now open to future exploits, but hey, that's your choice. They're not going to split patch lanes into "with vs without" mitigation--they'll all be with mitigation going forward. HT, likewise, is something that'll be left up to the users: enable or disable at your option. Intel has said that they don't feel it's necessary to entirely disable HT, but others of course are disagreeing with that. It may be that we see more Google-like disabled-by-default setups for servers and other security conscious systems, though.
|
|
#
?
May 17, 2019 16:27
|
|
|
DrDork posted:Except, as has already been noted, consumers aren't really much at risk, and the mitigations only minorly impact performance. For big data, where disabling HT might actually be painful, the cost of throwing out and replacing all your poo poo with enough AMD processors to get back to the same level of performance you had with Intel would be hilariously more expensive than just disabling HT / patching and buying a couple extra Intel boxes to make up the difference. If your workload parallelizes well, the higher end Epyc SKUs (7551 in particular) will already give you 2-3x more ops for your dollar over a similar Platinum Xeon. We're already moving and Zen2 only makes the advantage even greater with doubling the core density.
|
|
#
?
May 17, 2019 16:29
|
|
|
BangersInMyKnickers posted:If your workload parallelizes well, the higher end Epyc SKUs (7551 in particular) will already give you 2-3x more ops for your dollar over a similar Platinum Xeon. We're already moving and Zen2 only makes the advantage even greater with doubling the core density. Yeah, there certainly are applications for which AMD's hilarious core-per-dollar advantage over Intel makes for a compelling sales argument. But sometimes that isn't enough, especially when you're talking very large installs where the service, support, integration, tooling, etc., costs are often substantially more than the actual hardware costs. You can potentially have an AMD system that's 2x the cost/performance of Intel, and still end up going Intel because your tools already are designed for them, support contracts are more palatable, whatever. Regardless, that AMD is coming out swinging of late is nothing but a good thing for the CPU industry overall.
|
|
#
?
May 17, 2019 16:36
|
|
|
DrDork posted:Regardless, that AMD is coming out swinging of late is nothing but a good thing for the CPU industry overall.
|
|
#
?
May 17, 2019 16:37
|
|
|
Sorry, are you fabbing your own systems from scratch? The whole point of an OEM is to offload those to someone else.
|
|
#
?
May 17, 2019 16:38
|
|
|
Khorne posted:Intel is going to be forced to compete on price which is a win for everybody. They already are, the pricing is far more aggressive on the 62xx CPUs, and 42xx silvers are priced to be able to fight 2S vs AMD 1 socket.
|
|
#
?
May 17, 2019 16:44
|
|
|
Twerk from Home posted:They already are, the pricing is far more aggressive on the 62xx CPUs, and 42xx silvers are priced to be able to fight 2S vs AMD 1 socket. It's not far more aggressive, you get about 20% more CPU cycles for your dollar compared to the 1st gen scalable xeon. Intel Gold Xeons are somewhat competitive on pricing against the 7501 and 7601, but the 7551 is the obvious choice as its 5% less CPU for half the cost. For raw CPU cycles, you need Xeon platinum chips to compete with the huge quantity of cycles an Eypc socket gives you and you're paying a 4x price premium to do it. Yeah, your all core turbo frequency on Intel is going to be 15% higher but that doesn't mean much when you're getting 50% more cores on the AMD side.
|
|
#
?
May 17, 2019 17:14
|
|
|
DrDork posted:If you don't want the performance hit from the soft mitigation, you don't patch. That's it. You're now open to future exploits, but hey, that's your choice. They're not going to split patch lanes into "with vs without" mitigation--they'll all be with mitigation going forward. Kind of hard not to patch when software and hardware vendors cram patches down your throat by mandatory device updates.
|
|
#
?
May 18, 2019 08:07
|
|
|
Shipon posted:Kind of hard not to patch when software and hardware vendors cram patches down your throat by mandatory device updates. So read the documentation and adjust accordingly. Every single one of the mitigations that can have a performance impact gives you the ability to override and disable it on desktops, and they’re opt-in on servers.
|
|
#
?
May 18, 2019 14:20
|
|
|
Phoronix has the hard numbers, here's the tldr. https://www.phoronix.com/scan.php?page=article&item=mds-zombieload-mit&num=1 quote:If looking at the geometric mean for the tests run today, the Intel systems all saw about 16% lower performance out-of-the-box now with these default mitigations and obviously even lower if disabling Hyper Threading for maximum security. The two AMD systems tested saw a 3% performance hit with the default mitigations. While there are minor differences between the systems to consider, the mitigation impact is enough to draw the Core i7 8700K much closer to the Ryzen 7 2700X and the Core i9 7980XE to the Threadripper 2990WX.
|
|
#
?
May 18, 2019 23:57
|
|
|
Just bought a decommissioned hp z420 with an e5 1620 for 50 bucks. A drat good deal for essentially a 2600k, 16 gigs of ram, and what sounds like FIVE GODDAMN LEAF BLOWERS BEING FILTERED THROUGH A GNOMES WHISTLING ANUS AT IDLE. Also the firepro v4900 in it is busted so I'm not getting vidéo out but meh it boots. Good enough until I start switching poo poo over tomorrow. What cooler would be quiet and good enough for 4.2ish ghz? Just grab a hyper 212 off Craigslist for 15 bucks?
|
|
#
?
May 19, 2019 00:45
|
|
|
Yeah, basically any stand alone cooler should be fine. A 212 Evo is theoretically good up to about 180W, while a 1620 has a TDP of ~130W, though obviously a bit more if you plan on overclocking it.
|
|
#
?
May 19, 2019 00:53
|
|
|
MaxxBot posted:Phoronix has the hard numbers, here's the tldr. lol...but hey, users can disable these mitigations if they want to so no problemo
|
|
#
?
May 19, 2019 02:15
|
|
|
SeaGoatSupreme posted:Just bought a decommissioned hp z420 with an e5 1620 for 50 bucks. A drat good deal for essentially a 2600k, 16 gigs of ram, and what sounds like FIVE GODDAMN LEAF BLOWERS BEING FILTERED THROUGH A GNOMES WHISTLING ANUS AT IDLE. I have a z600 in my office and it's whisper quiet, try removing dust from fans with compressed air and go into the bios to check if the system fan is set to maximum.
|
|
#
?
May 19, 2019 06:47
|
|
|
SeaGoatSupreme posted:Just bought a decommissioned hp z420 with an e5 1620 for 50 bucks. A drat good deal for essentially a 2600k, 16 gigs of ram, and what sounds like FIVE GODDAMN LEAF BLOWERS BEING FILTERED THROUGH A GNOMES WHISTLING ANUS AT IDLE. FYI, I don't think that motherboard will let you OC. You may want to check the BIOS, there is a fan idle speed control there ("thermal policy" perhaps) and it might have been set to maximum. Or it might just need thermal paste. If you replace the fan, be aware of the possibility that HP may have used a non-standard fan mounting pattern. I have run into "issues" with HP doing proprietary stuff on the Z-series workstations before... the power supply on my Z400 is proprietary and uses a non-standard 24-pin connector.
|
|
#
?
May 19, 2019 06:48
|
|
|
Paul MaudDib posted:FYI, I don't think that motherboard will let you OC. I haven't bothered yet, but the motherboard should be inconsequential to software over clocking with xtu, right? Either way it *should* be better than what I have with regard to minimum framerates. I hope. The fan perpetrator was a 3 pin 60mm "memory fan". What the gently caress is that bullshit. Why
|
|
#
?
May 19, 2019 21:43
|
|
|
Someone at Intel is going to only conclude that the HT segmentation was a brilliant idea all along.
|
|
#
?
May 20, 2019 10:15
|
|
|
They're gonna start calling their non-HT lines security edition or some poo poo.
|
|
#
?
May 20, 2019 19:48
|
|
|
Deuce posted:They're gonna start calling their non-HT lines security edition or some poo poo. "Secure by Design(tm)"
|
|
#
?
May 20, 2019 21:48
|
|
|
Deuce posted:They're gonna start calling their non-HT lines security edition or some poo poo. 
|
|
#
?
May 22, 2019 21:26
|
|
|
Huawei allegedly leaks Intel CPU roadmap. Includes DDR5 and PCIe 5.0.
|
|
#
?
May 22, 2019 22:27
|
|
|
Burn those bridges good, Huawei!
|
|
#
?
May 22, 2019 22:30
|
|
|
Huawei, champion of the people once again.
|
|
#
?
May 22, 2019 22:48
|
|
|
First glance at the first presentation slide I thought it said "Increasing the price of innovation".
|
|
#
?
May 22, 2019 23:59
|
|
|
I'm not sure this is the right thread for this question, but I thought people here might be able to give insight. I'm planning to buy a desktop computer to replace my 2013 ThinkPad laptop in the next 1-2 years. I'm just starting to see some performance slowdown now, and my life circumstances have changed such that I don't need my primary computer to be a laptop anymore. I know that there's always something on the horizon and that you can't futureproof a computer, but I was curious whether there's anything really big with computer technology that people think will be on the market within the next couple of years and are really anticipating. I'm sure that every year the processors get 3% faster and that sort of thing; an example of what I mean is the introduction of SSDs where it was a really noticeable difference-maker over HDDs. Since I'm not in a rush to replace the laptop, I feel like I could wait for something big like that unless it seems several years away. If everything on the horizon is just incremental improvements, I'll probably upgrade sooner rather than later. Thanks for any knowledge folks can share! edit: for reference, my current laptop's processor is an Intel Core i5-3230M at 2.60 GHz, so I'm sure that current-market tech would be a big jump over what I've got now, especially going from laptop to desktop.
|
|
#
?
May 23, 2019 01:11
|
|
|
The big stuff is within your window. Zen2 in a few months and then Intel hopefully combining a significant process leap with some architectural improvements. After that it should all be incremental for a while. Even those are going to be doing well if they're like 10% steps forward. K8.0 fucked around with this message at 01:18 on May 23, 2019 |
|
#
?
May 23, 2019 01:15
|
|
|
Intel allegedly tried to pay university researchers a... uh, bug bounty that no one asked for. https://www.techpowerup.com/255563/intel-tried-to-bribe-dutch-university-to-suppress-knowledge-of-mds-vulnerability
|
|
#
?
May 23, 2019 07:31
|
|
|
eames posted:Intel allegedly tried to pay university researchers a... uh, bug bounty that no one asked for.
|
|
#
?
May 23, 2019 14:24
|
|
|
Khorne posted:That's actually pretty cool. I mean, it clearly benefits Intel but it also benefits researchers. No, it sucks because they're trying to manipulate the research and the amount they were offering to an entire research team is absolute peanuts compared to what they're going to be able to pull down in salary by doing the full disclosure and graduating with this under their belt.
|
|
#
?
May 23, 2019 15:57
|
|
|
BangersInMyKnickers posted:No, it sucks because they're trying to manipulate the research and the amount they were offering to an entire research team is absolute peanuts compared to what they're going to be able to pull down in salary by doing the full disclosure and graduating with this under their belt. Usually these things come with a 3-6 month NDA so the supplier can fix the problem in the public before the issue is made public. They still get the full disclosure and credit.
|
|
#
?
May 23, 2019 16:57
|
|
|
ufarn posted:Huawei allegedly leaks Intel CPU roadmap. Includes DDR5 and PCIe 5.0. I want to go and scrub the companies capable of making SerDes / transceivers that can run at those line rates for PCIe 5.0...has to be a short list. Then again, 10Gb/s is basically commoditized and even the weirdest, cheapest SoCs coming out sport PCIe 2.0 for high-speed interconnect.
|
|
#
?
May 23, 2019 17:17
|
|
|
karthun posted:Usually these things come with a 3-6 month NDA so the supplier can fix the problem in the public before the issue is made public. They still get the full disclosure and credit. Intel was already getting that, its par the course for responsible disclosure. Intel was trying to bribe them to make the impact look less severe. It was a poo poo move by a company caught with its pants down
|
|
#
?
May 23, 2019 17:19
|
|
|
K8.0 posted:The big stuff is within your window. Zen2 in a few months and then Intel hopefully combining a significant process leap with some architectural improvements. After that it should all be incremental for a while. Oh, great, sounds like it'll be a short wait then. One question: Zen2 is AMD, right? The last time I did a computer purchase, folks seemed to think that AMD was just a total joke. Have they managed to catch up or even leapfrog Intel over the last half-decade? edit: I'm a big dummy who totally forgot what was posted literally on the same page as this question, and who only remembered upon seeing it when hitting post. Ignore me! surf rock fucked around with this message at 17:43 on May 23, 2019 |
|
#
?
May 23, 2019 17:41
|
|
|
movax posted:I want to go and scrub the companies capable of making SerDes / transceivers that can run at those line rates for PCIe 5.0...has to be a short list. Then again, 10Gb/s is basically commoditized and even the weirdest, cheapest SoCs coming out sport PCIe 2.0 for high-speed interconnect. I’m shocked that they figure the current type CEM connectors are good enough for those rates. Maybe with a 2” trace (on megtron 6)!! Gen4 is going to be a real blink and you missed it spec sort of like what Gen2 was like. The big phy producers will be cadence and synopsys though, as usual.
|
|
#
?
May 23, 2019 17:43
|
|
|
|
| # ? May 20, 2024 04:48 |
|
|
surf rock posted:Oh, great, sounds like it'll be a short wait then. One question: Zen2 is AMD, right? The last time I did a computer purchase, folks seemed to think that AMD was just a total joke. Have they managed to catch up or even leapfrog Intel over the last half-decade? Buldozer was an absolute joke, but Zen is pretty good if you're will to trade off clock speed for more cores and Zen2 looks like its going to close the gap real fast
|
|
#
?
May 23, 2019 17:54
|
|