New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Comments
@PulsedMedia (and others)
Yes, intel is known to be, uhm, "creative" regarding the meaning of TDP, and yes. AMD's TDP numbers are more honest.
But my point was a different one. It was about having an array of low-power 2-8 cores products. sure AMD had some but they never received a lot of love and if you saw one, for example in a router, it virtually always was a CPU that was available only to board builders and in 1k or more quantities.
AMD seems to have gotten better at it; now they offer the "small Ryzens" for notebooks and even some quite low power versions (about 12 or 15W) - but still, those are but low core count Ryzens with a trimmed clock.
Now, look at the V series. Many never even heard about those. And it's not really easy to find a choice of boards with those.
But OK, I guess AMD wasn't strong enough to attack on all fronts and had to rip open one side of the intel castle first.
Yes, AMD did not make low power server products at all.
Used to buy many E-350 mITX boards tho, still use them, Atoms from same era are trash but the E350s consumed less energy, DDR3 upto 32Gigs if i recall right etc.
Typically, in my experience, AMD tends to max out at the TDP rating mentioned, Intel using 1.5-2x the mentioned TDP rating, and platform differences ofc too, but somehow it always ends up AMD being less power hungry. Hell, we use hundreds of old opterons to this date, newer xeons sucks just too much power to replace those opterons, and that's even without all these security issues! Security issues makes them damn near useless.
Now we also have those AsRock A300 based "minidedis", they are not supposed to be used as servers, but for low power high density, they work a treat! We made custom fan shrouds etc. and i am always stunned how little power they consume.
Cannot wait when we get to start racking our custom form factor servers with the AsRock Rack boards!
I knew why KASLR was developed. As far as I'm concerned, KASLR may be a nice thing to have, but not very useful. Leaking the kernel address is bad, but not serious, certainly not very serious.
SGX is a rubbish to begin with. It's a glorified DRM at best.
Oh, I see. All the kernel developers and quite a few scientists are all wrong and clueless because some guy in some forum said so, because apple is accused of having done a poor job, and because "naing" says so.
One might have a discussion on that and I even might agree to a large degree - but SGX is a decisive security device for many, it's about the best they have/had for some secure volatile storage.
I state that so clearly because sometimes a viewpoint seems to be sensible from a theoretical perspective, but many people who have to be practical have to live with what's available.
Did you look at the link? He provided a technical argument to back up his opinion and you're the "some guy in some forum said so". Feel free to quote "All the kernel developers and quite a few scientists" to counter his assertion. How can you not see the obvious that you're very bad at debating? Do we need another angstrom post calling you out? You still haven't learned how to use quotes, either.
Announcement:
I will from now on not any more inform about security relevant issues and events.
You can thank TimboJones.
And nothing of value was lost. If people relied on you for security relevant issues and events, they were already fucked in the first place. Your posts make poor arguments without providing data to back up your opinion. You assert the issue is "very serious", and then when someone disagrees, while providing very detailed reasons why you're mistaken, you have to go cry to mommy instead of technically engaging in robust discussion.
Jesus, so much valuable insight (sarcasm) from that paragraph that isn't relevant to the topic you started and it's a lot of words to say fuck all. What is your point from this? Security issues that are "very serious" are ok because of lower power (not quantified, no source, just your opinion)? That's just bad logic.
Lol, wut? You're the one who started a thread talking about how they lied about fixing the flaws that were disclosed to them while designing their latest gen chips. I haven't seen any neutral observer tone from you on Intel in any previous Intel security thread. This is just a bonkers statement.
I can't recall another person constantly blaming others for their faults other than orange man. If you have colleagues, I pity them.
@angstrom already clearly said what your problem is:
Print that quote out, read it every morning and then take some responsibility, FFS. Blaming me for not being able to have technical debate without throwing a hissy fit is pretty pathetic for a grown man. What a baby.
Lol, and we were planning to buy some xeon v3 decommisioned servers.
Guess we'll just go Epyc, some supermicro boards can fit in those old opteron 1u chassis fine, you just need a 20 to 24 pin adapter for the psu and remove the io shield.