ssp

14 ms·

91.
by ssp 15y ago
Overview of the new instructions (cutted and pasted from the PDF): AVX2 promotes the vast majority of 128-bit integer SIMD instruction sets to operate with 256-bit wide YMM registers. AVX2 instructions are encoded using the VEX prefix and
92.
New instructions in Intel's Haswell microarchitecture(software.intel.com)
37 points by ssp 15y ago|14 comments
93.
Intel Should Break Up(linleygroup.com)
2 points by ssp 15y ago|0 comments
94.
E3 2011: John Carmack on Rage, PC graphics, iOS games, and OnLive(reviews.cnet.com)
2 points by ssp 15y ago|0 comments
95.
by ssp 15y ago
Maybe I'm missing the joke, but Emacs was famous for bringing multi-megabyte machines to their knees and garbage collecting for minutes. The joke was that emacs stood for "eight megabytes and constantly swapping".
96.
by ssp 15y ago
Have you filed patents on it, then? I'd be interested in seeing the applications.
97.
by ssp 15y ago
A variant of this is to publicly display the logarithm of the score rounded up to an integer, maybe even as a number of stars. If the base-3 logarithm was used, the displayed scores would be: -infinity to 0: bad comment, should be
98.
by ssp 15y ago
It's not clear to me that "number of upvotes minus number of downvotes" is a meaningful measure of anything, whether its displayed or not. An upvote from a high-quality user should count for more than one from a low-quality user. If you con
99.
by ssp 15y ago
while happily plundering the fairy tales and characters of earlier centuries. It's even more striking if you consider that those fairy tales and characters are the products of many, many retellings and adaptations throughout history. The
100.
by ssp 15y ago
It makes writing a JIT so pleasant that I meant to write an article walking through the process of building a simple JIT. Please do, if you have the time. DynASM really is a stroke of genius. The way it works is that the JIT contains bits
101.
by ssp 15y ago
And, I think you're going to be very lonely at your meetings if you start a group pushing for the state to take over currently privately owned infrastructure. I'd show up. Nationalizing the communications hardware (the towers and cables),
102.
by ssp 15y ago
The binary heap is not Knuth's algorithm. He just analyzed it in great detail at a time where caches weren't as important as they are now. If Knuth were starting out today, I'm sure he would be interested in cache-aware algorithm analysis.
103.
Getting Vertical -- What NVIDIA's Acquisition of Icera Means(jonpeddie.com)
9 points by ssp 15y ago|1 comments
104.
Why Ten Bits is Better Than Eight(codingrelic.geekhold.com)
1 points by ssp 15y ago|0 comments
105.
Poor Old Nokia(electronicsweekly.com)
1 points by ssp 15y ago|0 comments
106.
Death of a Product Manager(crankypm.com)
53 points by ssp 15y ago|18 comments
107.
by ssp 15y ago
Part of this is probably Apple trolling Intel, but this one is pretty believable because Apple generally don't care about protecting their developers' investments, and because they can design ARM chips themselves.
108.
by ssp 15y ago
Interesting, thanks. I think there is a typo in the formula that follows "In this case our random variable is log(X). So we get:". On the third line, an n is missing after E(log(X)).
109.
by ssp 15y ago
Depends on what you mean by "general", I suppose, but if 'object' is a local variable in a Java-ish language, then there is a fairly straightforward data flow analysis that can do a good conservative estimate.
110.
by ssp 15y ago
The problem is that if the code looks like this: if (object.some_field != null) { some_unrelated_call(); object.some_field.foo(); } proving that some_unrelated_call() didn't set object.som
111.
by ssp 15y ago
The main problem with non-nullable types is that the static analysis involved doesn't work for object fields., so you can't write code like this: if (object.some_field != null) object.some_field.foo(); because t
112.
by ssp 15y ago
Yes, it's called sysprof-cli. It should be built and installed out of the box.
113.
by ssp 15y ago
Also take a look at sysprof: http://sysprof.com The 1.1.6 release uses the same kernel API as perf, but displays full call trees. Screenshot: http://sysprof.com/screen-shot-4.png
114.
by ssp 15y ago
Maybe there's no financial market, but it seems like a tremendous good. An Open Wifi protocol might be difficult to sell, but selling better routers should be possible. The current ones are ugly and difficult to use.
115.
by ssp 15y ago
It could have been intentional. Whoever designed the hardware might well have realized that it was important to be deliberately and obviously different from the iPhone, and that the various managers at Motorola/Verizon/Google just didn't ge
116.
Crazy for consumers? RIM is insane(ries.typepad.com)
2 points by ssp 16y ago|0 comments
117.
by ssp 16y ago
It could be. I always thought that an interesting way to compete with Google would be to index the web and then sell map/reduce access to that index.
118.
by ssp 16y ago
Indeed. They could consider doing something similar to Red Hat: http://www.redhat.com/legal/patent_policy.html
119.
by ssp 16y ago
Make a graph containing edges from each user to the comments they voted for, and from each comment its author. Then run something like PageRank on it and show the resulting ranks of both comments and users. It would help with comment qualit
120.
Emacs Lisp now has lexical bindings (lists.gnu.org)
14 points by ssp 16y ago|0 comments

More ›