How about until the agreed upon 1/14/2020 or whatever date in our fucking contracts?
They're blocking newer CPUs from accessing Windows Update and preventing them from downloading critical security patches. These patches do not require additional testing or development to work on PCs with the newer CPUs, and the newer CPUs do not magically make the gaping security holes go away.
Nobody wants more long and boring Tolkien shit. Remember the Hobbit trilogy? It failed.
The original trilogy sold about 73% more tickets (with a greater domestic share), made slightly more money ($3b 2001-2003 dollars vs $2.9b 2012-2014 dollars), and had less than half the budget. Ancillary licensing (toys, video games) was also much, much more lucrative.
notepad++ could not rely on external DLLs. Monolithic executables should make a comeback. Storage and memory are cheap, and we'll never see an end to the attacks that rely on manipulating shared memory. Using shared memory for anything important is like using a public bulletin board to file your taxes.
At the beginning of the second world war we were caught almost completely by surprise at Pearl Harbor in Hawaii when naval and air forces of the Empire of Japan wrecked our Pacific Fleet.
Fucking wrong. We were warned by the British well in advance. We let the attack happen so we could finally get public support for the war.
Does Windows 10 even support BIOS mobos that don't have partial/faked EFI support? Looking forward, everything is fucked. UEFI everywhere, Windows 10 everywhere, IME/PSP everywhere. BIOS implementations can be insecure as fuck too, but there's typically the option to lock the BIOS to prevent this. Modern enthusiast/gamer boards have a dual BIOS option which lets you toggle a good copy of the BIOS if the current one fucks up or you bork it with overclocking settings. Older MSI boards used to have 2 physical chips, but I'm fairly certain that the toggle is completely physical now, so you could still be fucked. Theoretically the toggle control would be very low level and represent a very small attack surface, so it shouldn't be hard to implement securely. But I haven't seen a modern gamer board with a physical jumper to write protect shit in a while, so while you have the option to toggle to a known-good BIOS, you'd only ever do that if you knew your current BIOS got fukt.
Secure boot doesn't protect shit beyond making you unable to boot to unsigned shit. The firmware is still as buggy and exploitable as fuck. And now wvery fuckign peripheral has firmware being talked to and running shit pre boot!
Are you retarded? You think UEFI presents a SMALLER attack surface? And you think that's why it boots faster? Hint: It doesn't boot faster. It may warm boot Windows faster, but that's got nothing to do with UEFI vs. BIOS. BIOS can achieve the same thing because the "fast booting" you're seeing is mostly due to skipping shit and not making devices available. Intel boards for example have 2 or 3 levels of fast booting, with warnings telling you that USB won't be available until OS handoff, you'll need to hold the button for 5 seconds or pull power to revert if you can't get into the OS, etc. It's akin to skipping POST and ignoring devices for boot / OROM consideration in BIOS.
The IME does some useful stuff like allowing you to VNC into your headless server's BIOS. It's far from useless for a lot of people, the issue is that you can't completely disable it.
I get BIOS over serial (over LAN) on old Dell servers because they have an IPMI chip (Dell calls it the BMC - baseboard management controller I believe). I can completely disable this if I want to. It doesn't need IME (at least older versions don't.) HP's equivalent is called "iLO" (integrated lights out). Not sure if that taps into IME.
It's too much code at too low a level, and it's too easy to manipulate. I for one would rather pay a nominal fee to have a new ROM chip sent to me. Remember when you could just pop those babies in and out? Remember when we had jumpers to protect and reset BIOS, boot sectors, etc.?
Yes, [U]EFI has good features and goes far beyond what BIOS can do, but so what? Outside of supporting hardware and booting to the point of OS handoff, the BIOS (either BIOS proper or [U]EFI) is supposed to be as minimal as possible. BIOS has been hacked to hell to support all sorts of shit like that at the behest of the various motheboard manufacturers. If we just had a newer BIOS developed by a central body that didn't try to completely reinvent the wheel as a helicopter, we'd be much better off.
The only functions of these things (Intel's is called the Management Engine) are backdoors and DRM. At the high end enterprise level they can be used for remote configuration and asset tracking, but:
They don't prevent theft. Despite bold claims, they don't actually result in recovering stolen shit either. I'm sure they have a handful of cases to point to, but recovery is rare. If you care about security you're using physical locks to keep things from walking away and encryption for when someone is determined.
No one remotely configures workstations at a low level. You buy them, image them, and hand them off. BIOS updates? Are you kidding me? For servers, various proprietary solutions exist built on top of open standards. Remote configuration is a big deal here, but we don't need an embedded, all-powerful black box to do it. The dumbest, cheapest (free) IPMI implementation can handle getting power status, rebooting, and piping BIOS over serial (and serial over LAN). And it can be turned the fuck off.
But AMD won't be removing it, so they could at least allow binary blobs to be loaded which disable functionality. (Or give us a config option or jumper to do the same.)
Asking them to open source the whole damn thing and hand over signing keys is asking for the moon. It would be great, sure. But I'd settle for the much more reasonable "disable to a fair degree of certainty" option.
No, this is not a setback for Bitcoin. It's a setback for some shitclowns who want to sell Bitcoin via an ETF. Bitcoin - the network, currency, and blockchain - are unaffected by this.
I've been reading and trolling Slashdot for quite some time. I don't recognize you or any such allegations you claim to have made.
Put up or shut up.
But beyond your little jerkoff session, of COURSE Apple can push code to devices without user intervention. Anyone who ever thought otherwise is a fool.
Hey, dumbass. I don't have a 7700k. I do have Windows (7), and an old i7-2600k. And if you had read my post, you'd realize that I fucking mentioned other workloads. I call out the 7700k and gaming performance specifically because that's the worst situation for Ryzen. I also point out the best situation for Ryzen.
The 1800X is a shitty choice for gaming. Perhaps drivers, coptimized software, and more mature motherboards/BIOSes will help over time. AMD's shit does tend to improve with age, but there's a huge delta there.
In general, the reviews are negative. We knew Ryzen would be behind in single-threaded performance, and it is. But it's also behind in multi-threaded performance in a lot of benchmarks for some reason. It beats out Intel's offerings in certain workloads (primarily video encoding), but it gets its ass handed to it in games.
There are huge discrepancies across the benchmarks and people aren't sure what to make of it yet. Do we need BIOS/microcode updates for shit like better memory clocks/timings? Does software need to be recompiled? Why does disabling SMT increase performance so drastically in so many workloads? When will we get a driver and Windows scheduler update to fix some of the performance issues related to core parking, etc.?
The chips have a hard temperature limit of only 75 degrees and have limited overclocking potential. Under any sort of sustained load they throttle fairly quickly, to boot.
The 1800X is basically useless unless you need a lot of cores for specific workloads. The 1700X looks to be a better all around choice, but even the lower price of this model doesn't justify choosing it over a i7 7700K for gamers.
I had hoped that Ryzen parts would be the clear winner in performance/$, but they're not unless you're doing software rendering/encoding, or running VMs.
I'm all for having things scripted or menuized, or otherwise made foolproof over manually keying things in that aren't security-sensitive (such as passwurdz). But triggering those scripts and running the "do this complex thing" job should have a human at the trigger, watching things as they burst into flames (or don't).
Another option for complex procedures is to have 2 people to serve as a check against each other. These kinds of checks are commonplace in the military and various regulated industries (mining, manufacturing, precision butt scratching). But they cost more, and companies hate that.
Uh, they traded commodities and got illegally scammed (flipping their own scam against them, yes).
Why would you short at $17? The shitty web 3.0 IPOs hit peak mindshare on day 1, not day 0. The people not in on the con have to wait until day 1 to initiate their trades. Only the big boys get their shit in in advance. Everyone else buys in after ASAP in the mad scramble. The bubble bursts when the big boys want to cash out or when everyone else gets tried of yet another quarter of no profit, no growth, and no dividends.
Slashdot Mirror
How about until the agreed upon 1/14/2020 or whatever date in our fucking contracts?
They're blocking newer CPUs from accessing Windows Update and preventing them from downloading critical security patches. These patches do not require additional testing or development to work on PCs with the newer CPUs, and the newer CPUs do not magically make the gaping security holes go away.
Nobody wants more long and boring Tolkien shit. Remember the Hobbit trilogy? It failed.
The original trilogy sold about 73% more tickets (with a greater domestic share), made slightly more money ($3b 2001-2003 dollars vs $2.9b 2012-2014 dollars), and had less than half the budget. Ancillary licensing (toys, video games) was also much, much more lucrative.
It's MightyMartian. No, he did not read it. At best he glanced at it, failed to understand it, and posted some bullshit anyway.
notepad++ could not rely on external DLLs. Monolithic executables should make a comeback. Storage and memory are cheap, and we'll never see an end to the attacks that rely on manipulating shared memory. Using shared memory for anything important is like using a public bulletin board to file your taxes.
At the beginning of the second world war we were caught almost completely by surprise at Pearl Harbor in Hawaii when naval and air forces of the Empire of Japan wrecked our Pacific Fleet.
Fucking wrong. We were warned by the British well in advance. We let the attack happen so we could finally get public support for the war.
Dump the chip? Or use EEPROM (they're all EEPROM now anyway) and write a trusted version yourself?
Does Windows 10 even support BIOS mobos that don't have partial/faked EFI support?
Looking forward, everything is fucked. UEFI everywhere, Windows 10 everywhere, IME/PSP everywhere.
BIOS implementations can be insecure as fuck too, but there's typically the option to lock the BIOS to prevent this. Modern enthusiast/gamer boards have a dual BIOS option which lets you toggle a good copy of the BIOS if the current one fucks up or you bork it with overclocking settings. Older MSI boards used to have 2 physical chips, but I'm fairly certain that the toggle is completely physical now, so you could still be fucked. Theoretically the toggle control would be very low level and represent a very small attack surface, so it shouldn't be hard to implement securely. But I haven't seen a modern gamer board with a physical jumper to write protect shit in a while, so while you have the option to toggle to a known-good BIOS, you'd only ever do that if you knew your current BIOS got fukt.
Secure boot doesn't protect shit beyond making you unable to boot to unsigned shit. The firmware is still as buggy and exploitable as fuck. And now wvery fuckign peripheral has firmware being talked to and running shit pre boot!
Are you retarded? You think UEFI presents a SMALLER attack surface? And you think that's why it boots faster? Hint: It doesn't boot faster. It may warm boot Windows faster, but that's got nothing to do with UEFI vs. BIOS. BIOS can achieve the same thing because the "fast booting" you're seeing is mostly due to skipping shit and not making devices available. Intel boards for example have 2 or 3 levels of fast booting, with warnings telling you that USB won't be available until OS handoff, you'll need to hold the button for 5 seconds or pull power to revert if you can't get into the OS, etc. It's akin to skipping POST and ignoring devices for boot / OROM consideration in BIOS.
The IME does some useful stuff like allowing you to VNC into your headless server's BIOS. It's far from useless for a lot of people, the issue is that you can't completely disable it.
I get BIOS over serial (over LAN) on old Dell servers because they have an IPMI chip (Dell calls it the BMC - baseboard management controller I believe). I can completely disable this if I want to. It doesn't need IME (at least older versions don't.)
HP's equivalent is called "iLO" (integrated lights out). Not sure if that taps into IME.
When will people admit that [U]EFI was a mistake?
It's too much code at too low a level, and it's too easy to manipulate. I for one would rather pay a nominal fee to have a new ROM chip sent to me. Remember when you could just pop those babies in and out? Remember when we had jumpers to protect and reset BIOS, boot sectors, etc.?
Yes, [U]EFI has good features and goes far beyond what BIOS can do, but so what? Outside of supporting hardware and booting to the point of OS handoff, the BIOS (either BIOS proper or [U]EFI) is supposed to be as minimal as possible. BIOS has been hacked to hell to support all sorts of shit like that at the behest of the various motheboard manufacturers. If we just had a newer BIOS developed by a central body that didn't try to completely reinvent the wheel as a helicopter, we'd be much better off.
Can someone explain why "Kelly boy" is used to refer to a girl (presumably one named Kelly)?
Ideally the thing wouldn't exist.
The only functions of these things (Intel's is called the Management Engine) are backdoors and DRM.
At the high end enterprise level they can be used for remote configuration and asset tracking, but:
They don't prevent theft. Despite bold claims, they don't actually result in recovering stolen shit either. I'm sure they have a handful of cases to point to, but recovery is rare. If you care about security you're using physical locks to keep things from walking away and encryption for when someone is determined.
No one remotely configures workstations at a low level. You buy them, image them, and hand them off. BIOS updates? Are you kidding me? For servers, various proprietary solutions exist built on top of open standards. Remote configuration is a big deal here, but we don't need an embedded, all-powerful black box to do it. The dumbest, cheapest (free) IPMI implementation can handle getting power status, rebooting, and piping BIOS over serial (and serial over LAN). And it can be turned the fuck off.
But AMD won't be removing it, so they could at least allow binary blobs to be loaded which disable functionality. (Or give us a config option or jumper to do the same.)
Asking them to open source the whole damn thing and hand over signing keys is asking for the moon. It would be great, sure. But I'd settle for the much more reasonable "disable to a fair degree of certainty" option.
No, this is not a setback for Bitcoin. It's a setback for some shitclowns who want to sell Bitcoin via an ETF.
Bitcoin - the network, currency, and blockchain - are unaffected by this.
Never use IBM or Oracle.
On time. On budget. Functional. Pick zero.
it's exceedingly unlikely that a larger bank is storing anything password related in plain text
LOOOOOOOOOOOOOOOOOOOOOOL
Where have you been for the past 40 years? Or even the past 4?
I've been reading and trolling Slashdot for quite some time. I don't recognize you or any such allegations you claim to have made.
Put up or shut up.
But beyond your little jerkoff session, of COURSE Apple can push code to devices without user intervention. Anyone who ever thought otherwise is a fool.
I also vote for KeePass. It's very nice and very extensible.
Ryzen supports ECC. Even on the desktop, if the mobo supports it.
Hey, dumbass. I don't have a 7700k. I do have Windows (7), and an old i7-2600k.
And if you had read my post, you'd realize that I fucking mentioned other workloads. I call out the 7700k and gaming performance specifically because that's the worst situation for Ryzen. I also point out the best situation for Ryzen.
$500 R7 1800X vs $340 i7 7700k.
http://www.gamersnexus.net/hwr...
The 1800X is a shitty choice for gaming. Perhaps drivers, coptimized software, and more mature motherboards/BIOSes will help over time. AMD's shit does tend to improve with age, but there's a huge delta there.
In general, the reviews are negative. We knew Ryzen would be behind in single-threaded performance, and it is. But it's also behind in multi-threaded performance in a lot of benchmarks for some reason. It beats out Intel's offerings in certain workloads (primarily video encoding), but it gets its ass handed to it in games.
There are huge discrepancies across the benchmarks and people aren't sure what to make of it yet. Do we need BIOS/microcode updates for shit like better memory clocks/timings? Does software need to be recompiled? Why does disabling SMT increase performance so drastically in so many workloads? When will we get a driver and Windows scheduler update to fix some of the performance issues related to core parking, etc.?
The chips have a hard temperature limit of only 75 degrees and have limited overclocking potential. Under any sort of sustained load they throttle fairly quickly, to boot.
The 1800X is basically useless unless you need a lot of cores for specific workloads. The 1700X looks to be a better all around choice, but even the lower price of this model doesn't justify choosing it over a i7 7700K for gamers.
I had hoped that Ryzen parts would be the clear winner in performance/$, but they're not unless you're doing software rendering/encoding, or running VMs.
I'm all for having things scripted or menuized, or otherwise made foolproof over manually keying things in that aren't security-sensitive (such as passwurdz).
But triggering those scripts and running the "do this complex thing" job should have a human at the trigger, watching things as they burst into flames (or don't).
Another option for complex procedures is to have 2 people to serve as a check against each other. These kinds of checks are commonplace in the military and various regulated industries (mining, manufacturing, precision butt scratching). But they cost more, and companies hate that.
Not much editor good.
Uh, they traded commodities and got illegally scammed (flipping their own scam against them, yes).
Why would you short at $17? The shitty web 3.0 IPOs hit peak mindshare on day 1, not day 0. The people not in on the con have to wait until day 1 to initiate their trades. Only the big boys get their shit in in advance. Everyone else buys in after ASAP in the mad scramble. The bubble bursts when the big boys want to cash out or when everyone else gets tried of yet another quarter of no profit, no growth, and no dividends.
As Megatron said to Optimus Prime as he shot him repeatedly (and fatally) while using Hot Rod as a shield: "FALL! FALL!"
The smart money will short this shit into the ground. Snapchat has fewer opportunities to generate revenue than fucking Twitter does.