Tuesday, August 27
Daily News Stuff 26 August 2019
Lazy Sunday Afternoon Edition
Lazy Sunday Afternoon Edition
Tech News
- The past and future of Zen. (Backplane)
AMD's new 7nm chips don't overclock very well, and likely never will. Intel will have the same problem when they reach 7nm, and already seem to be experiencing it at 10nm, which is why they're still announcing new 14nm chips.
5nm will just make things worse for overclockers. From here on in, your app is multithreaded, or it's toast.
- A deep dive on how to clean up the Ubuntu message of the day. (Bityard)
When something that simple requires a deep dive, maybe it needs to be rethought, or alternately, nuked from orbit.
- Moscow's brand new blockchain-based cryptographically secured voting system can be hacked in 20 minutes using an Apple IIe with a 48k RAM expansion. (ZDNet)
Well, not quite, and US voting machines aren't really any better.
- Node.js is worse than you think. (Reddit)
No matter how bad you think it is, it's worse than that.
Disclaimer: Komi still can't communicate. I'm now up to chapter 150 with just 64 more to go. Fortunately this is the anti-Berserk and the author is cranking out a new chapter a week.
Posted by: Pixy Misa at
12:40 AM
| Comments (3)
| Add Comment
| Trackbacks (Suck)
Post contains 199 words, total size 2 kb.
1
"AMD's new 7nm chips don't overclock very well, and likely never will."
I keep hearing that AMD's chips (and Intel and nVidia to a lesser extent, but everyone else as well) are released pretty much right at the elbow of the curve where to gain more performance, power usage goes through the roof. I also keep hearing that 14nm and smaller nodes are low-power-optimized processes. I will admit I know next to nothing about this but it seems like one of the foundries should be coming out with a process that is optimized for higher power. (Maybe it's not possible at this size but I haven't actually seen anyone say it's not.)
I keep hearing that AMD's chips (and Intel and nVidia to a lesser extent, but everyone else as well) are released pretty much right at the elbow of the curve where to gain more performance, power usage goes through the roof. I also keep hearing that 14nm and smaller nodes are low-power-optimized processes. I will admit I know next to nothing about this but it seems like one of the foundries should be coming out with a process that is optimized for higher power. (Maybe it's not possible at this size but I haven't actually seen anyone say it's not.)
Posted by: Rick C at Tuesday, August 27 2019 01:57 AM (Iwkd4)
2
A few years ago, someone on The Daily WTF wrote a chatbot in node, specifically because node is cancer.
I looked at that isBuffer package. I barely know node and my JS knowledge is way out of date but it looks like the isBuffer function will be fooled by any object that has a method called isBuffer. (I don't know why you would have a legit reason to do that OTTOMH.)
I looked at that isBuffer package. I barely know node and my JS knowledge is way out of date but it looks like the isBuffer function will be fooled by any object that has a method called isBuffer. (I don't know why you would have a legit reason to do that OTTOMH.)
Posted by: Rick C at Tuesday, August 27 2019 03:09 AM (Iwkd4)
3
Re: overclock capability. The fact that you can't do it is a bit disappointing but the fact that AMD's giving you the entire capacity of the chip is pretty nice.
Unless you got a dud sample that can't reach the rated boost clock, like I did. (Or hits 90C on a Prime95 Small FFT...with a 360MM radiator. Ugh.)
Unless you got a dud sample that can't reach the rated boost clock, like I did. (Or hits 90C on a Prime95 Small FFT...with a 360MM radiator. Ugh.)
Posted by: Rick C at Tuesday, August 27 2019 09:49 AM (Iwkd4)
50kb generated in CPU 0.0146, elapsed 0.1037 seconds.
58 queries taking 0.0945 seconds, 342 records returned.
Powered by Minx 1.1.6c-pink.
58 queries taking 0.0945 seconds, 342 records returned.
Powered by Minx 1.1.6c-pink.
