Yesterday, we reported on an article about the demise of the Alpha. That article was the first part in a series about the future of processor design. Today, part II has been published: "In terms of the architecture itself, AMD's Athlon 64 platform, at the stage it is at right now, does not offer that much of a performance advantage, and AMD should not be resting on its laurels. This is because on the desktop, interconnects as such play less of a role. It's on servers and multi-processing systems that you can take advantage of scaling, and that's where interconnects such as HyperTransport have a role. But when you talk about a single-chip desktop system, whether it's one, two or four cores, the efficiency of the chipset still plays a very important role."
To view parent comment, click here.
To read all comments associated with this story, please click here.
Member since:2005-07-06
Does SSE3 has a multiply-accumulate operation?
I remember that it was missing in previous version.
Little endian simplifies some assembly language trick, big endian simplifies other assembly language trick, I'm not sure that there is really a benefit of little endian for the assembly language programming here, but human read in big endian and this is a big difference.
What do you say that Thumb sucked? Because of the performance impact?
Anyway I've not said Thumb but Thumb2 where you can mix 32 and 16 bit instructions.
Member since:2005-07-06
Little endian simplifies some assembly language trick, big endian simplifies other assembly language trick, I'm not sure that there is really a benefit of little endian for the assembly language programming here, but human read in big endian and this is a big difference.
OMG, I can't believe people are still beating that particular dead horse.
The differences between the two approaches are so small that the choice is really only a matter of taste, and that's of course why we ended up with the industry split down the middle on that one.
The supposed advantages of either approach are laughable compared to the huge disadvantage of having to deal with both of them in heterogenous settings.
Member since:2005-07-06
Does SSE3 has a multiply-accumulate operation?
No, because x86 does not allow for three-operand instructions. There is a multiply-add instruction for two 16-bit integer vectors, but that's not the same.
Therefore multiply-accumulate requires separate multiply and add instructions. Whether that actually has a performance impact depends on what execution units there are and how the operations are scheduled. Anyone got experience with that?
"Ilja van Sprundel, a security researcher with IOActive, has discovered a large number of issues in the way various X client libraries handle the responses they receive from servers, and has worked with X.Org's security team to analyze, confirm, and fix these issues."
"Google is facing a new antitrust probe by the U.S. Federal Trade Commission into whether the company is using its leadership in the online display-advertising market to illegally curb competition, people familiar with the matter said." Sucks to be Google today, apparently.
"In the midst of the major press blitz surrounding its annual I/O Conference, Google dropped some unfortunate news about its instant messaging plans. In several places around the web, the company is replacing the existing 'Talk' platform with a new one called 'Hangouts' that sharply diminishes support for the open messaging protocol known as XMPP (or sometimes informally Jabber), and also removes the option to disable the archiving of all chat communications. These changes represent a switch from open protocols to proprietary ones, and a clear step backward for many users." That's why I always say: only suckers trust companies.
"The wave of departures is a sign of internal struggle, but amidst all the bad news, there is a glimmer of hope. According to an anonymous HTC executive cited by WSJ, the crucial HTC One is selling 'pretty good so far'. The executive revealed that HTC sold about five million One units since launch, a performance that is deemed encouraging. However, sales have been limited by the persisting supply issues that still plague the production of the HTC One." Finally some good news for HTC. Five million units amidst supply constraints is very impressive.
"Sony gave the PS4 50% more raw shader performance, plain and simple (768 SPs @ 800MHz vs. 1152 SPs & 800MHz). Unlike last generation, you don't need to be some sort of Jedi to extract the PS4's potential here. The Xbox One and PS4 architectures are quite similar, Sony just has more hardware under the hood. Weâll have to wait and see how this hardware delta gets exposed in games over time, but the gap is definitely there. The funny thing about game consoles is that itâs usually the lowest common denominator that determines the bulk of the experience across all platforms. On the plus side, the Xbox One should enjoy better power/thermal characteristics compared to the PlayStation 4. Even compared to the Xbox 360 we should see improvement in many use cases thanks to modern power management techniques." AnandTech does its usual in-depth thing."The Verge has learned that HTC's Chief Product Officer, Kouji Kodera, left the company last week. Kodera was responsible for HTC's overall product strategy, which makes the departure especially notable on the heels of the global launch of the make-or-break One. It's not just Kodera. In the past three-odd months, HTC has lost a number of employees in rapid succession." I really hope HTC pulls it together.
"It is with huge pleasure that the Debian GNU/Hurd team announces the release of Debian GNU/Hurd 2013. This is a snapshot of Debian 'sid' at the time of the Debian 'wheezy' release (May 2013), so it is mostly based on the same sources. It is not an official Debian release, but it is an official Debian GNU/Hurd port release." Important note: 75% of Debian packages are supported bu Debian GNU/Hurd. Impressive.
"The first killer app was VisiCalc. This early spreadsheet turned the Apple II from a hobbyist toy to a business computer. VisiCalc came with room for improvement, though. In addition, a new architecture and operating system, the Intel-based IBM PC and MS-DOS, also needed a spreadsheet to be taken seriously. That spreadsheet, released in early 1983, would be Lotus 1-2-3, and it would change the world. It became the PC's killer app, and the world would never be the same. On May 14, IBM quietly announced the end of the road for 1-2-3, along with Lotus Organizer and the Lotus SmartSuite office suite. Lotus 1-2-3's day is done." Impressive 30 year run."There's a disconnect between how Apple CEO Tim Cook sees his company's tax strategies and how some members of the US Senate view it. That became clearer than ever today after Cook and two other Apple executives testified before Congress, explaining why they're holding most of their international income in Irish subsidiaries like Apple Operations International, which declare no tax residency anywhere in the world. AOI hasn't filed a tax return anywhere in the world for the last five years, yet it earned $30 billion in income from 2009 to 2012, according to a Senate report released yesterday." Just because something is legal, doesn't mean it's just. Apple, Google, Microsoft, and many more - these companies might not be breaking any laws, but it's obvious to anyone that what they are doing is scummy.
The new WebKit Opera for Android has been released. "Opera 14 for Android is built on top of Chromium 26, with a total overhaul of the UI in native code, making it fit well with the latest Android design guidelines. Go get the build from Google Play or point your browser to m.opera.com, and give it a spin!"
Member since:
2005-08-22
I disagree: x86 is bad, and little crufty is understating: compare SSE vs Altivec for example.
SSE wasn't that great, but SSE2 and SSE3 are pretty good. AltiVec seems to be slightly faster, but SSE3 is more flexible in both operations and datatypes.
From ease-of-use point of view: its use of low-endian makes hard to read hex dump of memory, low number of register, not-orthogonal ISA...
Big endian makes it easier to stare at a random memory dump and figure out what's going on. Little endian makes code simplier as it makes it simplier to do things like treating an int as a short or a char. It's not something you'd directly notice unless you program in assembly, but it does tend to result in smaller code.
x86 instruction density can be matched by RISCs as ARM's Thumb2 have shown (Thumb had a performance impact, I don't know if it's still the case with Thumb2).
I don't know anything about Thumb2, but Thumb certainly sucked. The only time it made any sense was if your hardware couldn't provide a 32 bit memory interface, and even then you were still better off using ARM for certain things.