Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.
That seems like a mistake to me.
Intel's not in a good position for a massive multi-year R&D effort, even ignoring the risk of the current bubble popping in a few years and them entering the market just in time for the ram supercycle to enter it's oversupply and crashing prices phase.
Even 15ish years ago that was what my security conscious employer did for international travel. Your normal laptop stayed at home, you were issued a loaner to use instead. To the maximum extent possible it was was setup to be sterile on the way out and to use cloud self hosted web services...
I couldn't afford to replace my 3080 when I bought parts to refresh the rest of my system starting late last year.
Watching prices continue to soar I'm really hoping that card has at least a few more years in it. (And also regretting not getting a 4080 when they came out because the covid...
That was a factor. The second major one was from the other end, IIRC most IGPs of the time were relatively much smaller vs the CPU or chipset die (they existed because the minimum size of the chip due to the space needed for IO pins was less than the amount of die area needed for all the other...
Chipset IGPs have if anything even worse performance limitations than CPU ones. They're limited to standard DDR ram, but today the latency is lower for something on the CPU. They're also more thermally limited since there isn't space for a big cooler and fan.
To make one that's competitive...
IGP graphics are the modern equivalent of chipset ones. They share the same big limit in ram bandwidth; but probably have more thermal headroom since chipsets never had more than tiny heatsinks and fans which are dwarfed by even the smallest and crappiest of OEM CPU coolers.
I'd need to dig through a few months of old discord messages. He first mentioned them a few months ago, unless I'm misremembering Emil gave the impression that he was in talks with someone specific (not if you build a prototype they will come) as long as the prototypes are good everything would...
What happened was that in the past they could and frequently did lie by deception by advertising $XX and not mentioning $Y in double secret probation fees until you got your bill. Rules were changed (IIRC during the Biden admin) to require the label showing a breakdown of all the stupid fees...
Emil's savings are apparently depleted and he took a $1k/mo hit changing jobs.
Don't know to the first. At least some cases (not sure ifall) are supposedly done; with the last major delay being the previous planned manufacturing partner using "but they're in inches" as an excuse to back out.
Will it though? A lot of data center parts aren't in standard pcie card form factor. Even if a cheap adapter is available (and nothing about pcie5 signalling is cheap); will they even have 1 video out like some mining only cards from a decade ago or nothing at all?
I'm very skeptical of it lasting more than about 5 years, and wouldn't be surprised if it ends a year or two sooner.
The decade of shortages prediction is from equally long number goes up predictions for building AI data centers. The global economy isn't big enough to sustain that even if...
Latest update is that Emil thinks it will take about 10 months to save up the $5k for the prototypes the (potential?) outside investors want to see before moving forward.
He's reluctant to do any sort of crowd funding due to fear that if any other delays hit he'd be accused of scamming...
AWS low cost T2, T3, T3A, and T4G instances do share CPU resources. The baseline is 5-40% CPU load depending on tier.
https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/burstable-credits-baseline-concepts.html
The higher end instances might be able to just scale the number of cores being...
IIRC (been a few years since I needed to worry about it for work), everything except the absolute cheapest AWS tiers gave you dedicated cores not shared ones. IIRC the shared code option one gave you use credits equal to something like 25% load for your nominal count with the option to buy...