or Connect
AppleInsider › Forums › Mobile › iPad › Samsung chooses Intel CPU for new iPad-competing Android Galaxy Tab - report
New Posts  All Forums:Forum Nav:

Samsung chooses Intel CPU for new iPad-competing Android Galaxy Tab - report - Page 2

post #41 of 88
Quote:
Originally Posted by Gatorguy View Post

. . . and?
Would it then be using emulation or virtualization? I'm way over my head when it comes to technical issues like that and so would defer to you.

That still depends on what the piggybacking OS is built on. Let's say it's Windows 8 as the primary OS but you want to run Android 2.3 on it. That would be emulated if the only binary available is ARM-based. What version of Android are they running on it? If it's a newer version that runs on Intel than it would be virtualized and not only lose very little performance as opposed to be virtualized but also be much, much faster than any ARM version (assuming that they did a good job on that version of Android for x86/x86-64 and the virtualization layer is good. Considering both of these in 2013 I'd say Android on Haswell will look impressive.

So lets assume the best for Android OS. What about the apps? Can apps be compiled for both ARM and Intel as fat binaries? I ave absolutely no idea how Java works here. I'd think only the Java engine needs to be compiled for this, not the apps that run in Java, but I honestly have no idea.


"Your mama's so fat her binaries support multiple architectures."

"The real haunted empire?  It's the New York Times." ~SockRolid

"There is no rule that says the best phones must have the largest screen." ~RoundaboutNow

Reply

"The real haunted empire?  It's the New York Times." ~SockRolid

"There is no rule that says the best phones must have the largest screen." ~RoundaboutNow

Reply
post #42 of 88
Quote:
Originally Posted by Gatorguy View Post

. . . and?
Would it then be using emulation or virtualization? I'm way over my head when it comes to technical issues like that and so would defer to you.

Don't do it guys, it's a trap!

Virtual machines recompile bytecode into native machine code. Virtualization doesn't affect the application binary at all, it is used to abstract hardware to an OS in order to run native binaries. Emulation has to recreate hardware in software.

Adobe Flash is an example of a virtual machine - that's why Flash animations work on all hardware. It doesn't emulate or virtualize an OS like say VMWare. It takes Adobe's partially compiled bytecode and then recompiles it into native code.

The advantage is being able to target multiple architectures easily but there is a performance hit (I think estimates are typically around 10% hit - possibly why Android is laggy but probably not, most likely just because it's badly optimized for the hardware) and because authored software is only partially compiled, it can be reverse-engineered and stolen:

http://geeknizer.com/decompile-reverse-engineer-android-apk/

Think about that if you publish apps on Android. Someone could reverse-engineer your work, modify it and sell it as their own app. It's a wonder large publishers even publish on Android. They'd be as well making them open source:

http://insecurety.net/?p=641
post #43 of 88
Quote:
Originally Posted by malax View Post

That's interesting. I'm sure the Samsung semiconductor head isn't happy to lose business to Intel. Intel must have made a compelling case.

Clover Trail is an interesting product, intel literal threw the book out the window and started over with the design. Clover trail is SIGNIFICANTLY different than previous ATOMs.

Now that still doesn't mean that this new chip will cut it in a tablet, only time will tell there. It just means that Clover Trail isn't a complete failure all around like previous ATOMs. Lets face it AMDs low end APUs ran circles around previous ATOMs. With Clover Trail I suspect the results will be tied to Samsung's ability to optimize for the chip. This means getting the Java rip off to work well and to get proper native development working on the chip.

In any event what this does indicate is that Apple needs to really work hard on a high performance 64 bit chip. Frankly all those AMD engineers they recently hired are likely working on a project too far out. This will be most interesting.
post #44 of 88
Quote:
Originally Posted by SolipsismX View Post

That still depends on what the piggybacking OS is built on. Let's say it's Windows 8 as the primary OS but you want to run Android 2.3 on it. That would be emulated if the only binary available is ARM-based. What version of Android are they running on it? If it's a newer version that runs on Intel than it would be virtualized and not only lose very little performance as opposed to be virtualized but also be much, much faster than any ARM version (assuming that they did a good job on that version of Android for x86/x86-64 and the virtualization layer is good. Considering both of these in 2013 I'd say Android on Haswell will look impressive.

According to the story there's no Windows there at all. Acer is specifically avoiding it. So it would be using virtualization then? Thanks Soli. That was much easier to understand.

BTW there's mention of an x86 gaming project here, again with no mention of any OS other than Android.
http://www.zdnet.com/android-gaming-on-a-x86-powered-pc-with-iconsole-tv-7000015336/
melior diabolus quem scies
Reply
melior diabolus quem scies
Reply
post #45 of 88
Quote:
Originally Posted by dreyfus2 View Post

Hm, as was said before, running Android on an x86 architecture would be quite a surprise. I do not quite see Google porting Android to x86 at this point. Most of their tablet apps are scaled up phone apps already. Making it even more difficult does not make any sense.

Could that be a "Surface" competitor running Chrome OS instead?

It really isn't a big surprise, Android and Linux underneath are designed to run on a variety of hardware. I86 does give them more operating system flexibility though. As for "porting" much of Android runs on a Java virtual machine rip off.
post #46 of 88
Java code runs on a virtual machine in most cases and has since its development.
Quote:
Originally Posted by SolipsismX View Post

jragosta is correct.

Running MS Windows on a PPC Mac is emulation because Windows only had binaries for x86. Running MS Windows on an Intel Mac is virtualization because Windows understands the HW. That means it al had to be emulated so that the processor code understand what the OS was requesting. This made it very slow even on the fastest systems. Many Intel chips are even designed to allow a VM to take advantage of the HW with less overhead but the OS needs to understand x86 or x86_64 HW for this to work.
post #47 of 88
Quote:
Originally Posted by Gatorguy View Post

According to the story there's no Windows there at all. Acer is specifically avoiding it. So it would be using virtualization then? Thanks Soli. That was much easier to understand.

BTW there's mention of an x86 gaming project here, again with no mention of any OS other than Android.
http://www.zdnet.com/android-gaming-on-a-x86-powered-pc-with-iconsole-tv-7000015336/

Without knowing anything about this Asus AIO I wouldn't know. Are you saying that Android would be only OS installed on it? In that case it's running natively, assuming it's an Intel version of the Android.

"The real haunted empire?  It's the New York Times." ~SockRolid

"There is no rule that says the best phones must have the largest screen." ~RoundaboutNow

Reply

"The real haunted empire?  It's the New York Times." ~SockRolid

"There is no rule that says the best phones must have the largest screen." ~RoundaboutNow

Reply
post #48 of 88
Quote:
Originally Posted by SolipsismX View Post

Without knowing anything about this Asus AIO I wouldn't know. Are you saying that Android would be only OS installed on it? In that case it's running natively, assuming it's an Intel version of the Android.

Correct. Reportedly Acer has decided to avoid Microsoft altogether on this particular AIO.
melior diabolus quem scies
Reply
melior diabolus quem scies
Reply
post #49 of 88
Quote:
Originally Posted by Apple ][ View Post

I see talk of virtualization. Is that one of the many reasons as to why Android runs like complete shit?

Virtualization isn't exactly the same thing as running a Java Virtual Machine. I won't go into it here but rather will say research this in depth because this thread is already filled with misleading information. As to your question many things involved in the design of Android make it sluggish.
post #50 of 88
Quote:
Originally Posted by Gatorguy View Post

Correct. Reportedly Acer has decided to avoid Microsoft altogether on this particular AIO.

Edit: So native then. See, I told you I had no idea when it came to technical architecture questions. Thanks once again.
melior diabolus quem scies
Reply
melior diabolus quem scies
Reply
post #51 of 88
Quote:
Originally Posted by wizard69 View Post

Java code runs on a virtual machine in most cases and has since its development.

True, but I was trying to avoid Java 1) because I know little about it and 2) its virtualization seems be hard to grasp since the demarcation point is less obvious. With VMWare, Parallels, XenCenter, etc. the demarcation point is the kernel of the guest OS.



With JVM I guess it's the Java byte-code which talks to a Java engine so that any Java apps can run on any HW and OS that the the engine was for. The "build once" concept is great in theory but the resulting "build once, debug everywhere" reality always turned me off to liking it.

"The real haunted empire?  It's the New York Times." ~SockRolid

"There is no rule that says the best phones must have the largest screen." ~RoundaboutNow

Reply

"The real haunted empire?  It's the New York Times." ~SockRolid

"There is no rule that says the best phones must have the largest screen." ~RoundaboutNow

Reply
post #52 of 88
Quote:
Originally Posted by dreyfus2 View Post

What is the actual benefit of a x86 CPU when running Android? A product needs at least one selling point. It won't be lighter, it won't be cooler, it can't be cheaper, and it definitely won't have more battery life... so, what is the point?

 

You can put an "Intel Inside" sticker on it.

post #53 of 88
Quote:
Originally Posted by Gatorguy View Post

According to the story there's no Windows there at all. Acer is specifically avoiding it. So it would be using virtualization then? Thanks Soli. That was much easier to understand.
Something that is easy to understand doesn't imply correctness!

A Virtual Machine is a different concept than Virtualization. Marvin did a fairly good explanation above but I'm not sure I could clarify the concepts better in a post here so I will leave it alone.

As for Acer and Samsung with their i86 Androids I would want to even guess at what is native code and what is running on the Java/Dalvik Virtual Machine. Parts of Android could be improved with Native code but much of it wouldn't matter.
Quote:
BTW there's mention of an x86 gaming project here, again with no mention of any OS other than Android.
http://www.zdnet.com/android-gaming-on-a-x86-powered-pc-with-iconsole-tv-7000015336/

Not to confuse you even more but it isn't impossible to have Android, its operating system and its Virtual Machine running under virtualization on another platform.
post #54 of 88
Originally Posted by AppleInsider View Post

Samsung's decision to use an Intel chip for an Android device was portrayed as a "coup" for Intel in the report.

 

It's a coup for Intel, yes, because they finally have a toe-hold (fingernail-hold ?) on the post-PC era.

And it's a coup for coffee and tea drinking Android device users.  

They can set their coffee or tea cup on their Intel-based Samsung Galaxy Tab to keep their beverage piping hot.

Sent from my iPhone Simulator

Reply

Sent from my iPhone Simulator

Reply
post #55 of 88
Quote:
Originally Posted by wizard69 View Post

A Virtual Machine is a different concept than Virtualization.

True, but VM has different meanings depending on the context. I'd classify it as a polyseme: same spelling, same pronunciation, but different albeit related meanings.

"The real haunted empire?  It's the New York Times." ~SockRolid

"There is no rule that says the best phones must have the largest screen." ~RoundaboutNow

Reply

"The real haunted empire?  It's the New York Times." ~SockRolid

"There is no rule that says the best phones must have the largest screen." ~RoundaboutNow

Reply
post #56 of 88
Quote:
Originally Posted by SolipsismX View Post

True, but I was trying to avoid Java 1) because I know little about it and 2) its virtualization seems be hard to grasp since the demarcation point is less obvious. With VMWare, Parallels, XenCenter, etc. the demarcation point is the kernel of the guest OS.
Yeah this can be very confusing to try to explain over the net. This is especially the case when the terms sound alike to un informed ears. You mention a couple of pieces of software that virtualize hardware above to allow running guest operating systems, this is Virtualization. Those guest operating systems could then be running a Java app which runs in a Java Virtual Machine that runs on top of that Virtualized operating system.

Without confusing people more, the Java Virtual Machine emulates and idealized processor that executes Java byte code. In other words Java programs aren't compiled (in most cases) to target an underlying hardware processor but rather they are compiled to target a virtual processor. That processor is emulated by the Java Virtual Machine. I will ignore JIT compiling as that throws a curve ball at people.

I hope this doesn't confuse people even more.
Quote:


With JVM I guess it's the Java byte-code which talks to a Java engine so that any Java apps can run on any HW and OS that the the engine was for. The "build once" concept is great in theory but the resulting "build once, debug everywhere" reality always turned me off to liking it.

Yep. I never liked it either. There are some great apps out there written in Java so I'm not trying to dismiss it. Rather it just rubs name the wrong way.
post #57 of 88
Quote:
Originally Posted by wizard69 View Post

There are some great apps out there written in Java so I'm not trying to dismiss it. Rather it just rubs name the wrong way.

The only app I use that I think is even Java-based is a Pearson-Vue app for practice exams. That looks alright. The UI doesn't appear to be created in Java. I wish it was as I assume that would make it easier to port to Mac OS.

I had thought that Wireshark uses Java since it's so incredibly ugly and clunky but Wikipedia says the UI is GIMP. I wish there was a native app for Mac OS X (or Windows) that can do what Wireshark can do.

"The real haunted empire?  It's the New York Times." ~SockRolid

"There is no rule that says the best phones must have the largest screen." ~RoundaboutNow

Reply

"The real haunted empire?  It's the New York Times." ~SockRolid

"There is no rule that says the best phones must have the largest screen." ~RoundaboutNow

Reply
post #58 of 88
Quote:
Originally Posted by SolipsismX View Post

True, but VM has different meanings depending on the context. I'd classify it as a polyseme: same spelling, same pronunciation, but different albeit related meanings.

Yes the usage of the terms is at times very loose. Virtualization is more specific in most usage though. It is unfortunate that the industry doesn't have a better grip on this as it can be very confusing. The usage of the phrase Java Virtual Machine (JVM) though becomes very specific.

JVM implies the emulation of an abstract machine defined buy Sun to support the Java programming language. When you see the term Virtual Machine, that is without the Java qualifier, the definition becomes much looser. A virtual machine could imply something offered up by virtualization software that allows multiple operating systems to exist on one piece of hardware or it could imply an abstract machine other than Java.

So when one sees "Java Virtual Machine" any ambiguity should be gone. It is also why the preferred abbreviation is JVM to reduce confusion.
post #59 of 88
Quote:
Originally Posted by wizard69 View Post

It is unfortunate that the industry doesn't have a better grip on this as it can be very confusing.

I vote for the BASE-2 and BASE-10 for kilo, mega-, gigi-, et al. be addressed first. Using both 2^10 and 10^3 for kilo- is absurd. BASE-10 was first, and by a large margin, so I think the IEC Binary Prefixes of kebi-, mebi-, gibi-, et al. be utilized... but it might be easier to convince Americans to use the metric system for something other than 2 Liters of soda.
Edited by SolipsismX - 5/31/13 at 11:38am

"The real haunted empire?  It's the New York Times." ~SockRolid

"There is no rule that says the best phones must have the largest screen." ~RoundaboutNow

Reply

"The real haunted empire?  It's the New York Times." ~SockRolid

"There is no rule that says the best phones must have the largest screen." ~RoundaboutNow

Reply
post #60 of 88
Quote:
Originally Posted by SolipsismX View Post


I vote for the BASE-2 and BASE-10 for kilo, mega-, gigi-, tera-, et al. be addressed first. Using both 2^10 and 10^3 for kilo- is absurd. BASE-10 was first, and by a large margin, so I think the IEC Binary Prefixes of kebi-, mebi-, gibi-, tebi-, et al. be utilized... but it might be easier to get Americans to use the metric system for something other than 2 Liters of soda.

 

You'll pry my Imperial Measurements from my cold, dead fingers! :)

post #61 of 88
Quote:
Originally Posted by SolipsismX View Post

The only app I use that I think is even Java-based is a Pearson-Vue app for practice exams. That looks alright. The UI doesn't appear to be created in Java. I wish it was as I assume that would make it easier to port to Mac OS.
There are a few apps out there that are really impressive and written in Java. Eclipse is one of them even though it is a fairly buggy piece of software. Again I'm not dissing Java or Eclipse here as I use Eclipse often. I can't remember the name but there is a fantastic little Timing Diagram piece of software out there written in Java. There is also a text editor that I never liked.
Quote:
I had thought that Wireshark uses Java since it's so incredibly ugly and clunky but Wikipedia says the UI is GIMP. I wish there was a native app for Mac OS X (or Windows) that can do what Wireshark can do.

Actually I thought somebody was working on a native Mac OS interface to Wireshark. You might want to do a quick search though maybe the project died. I have to agree though Wireshark has one ugly interface.

Not to pull this thread completely off the rails but a few comment about Eclipse. One thing to like about it is that it is cross platform. This makes it very handy if you work on multiple platforms. Being Java it is very multi threaded and as such you really need a multi core processor to get Eclipse to work half decently, this happens to be true with a lot of Java apps. Fast multi core hardware has made Eclipse more bearable but native apps like XCode and Visual studio though work much better on their respective platforms. Even EMACS can feel snappier than Eclipse. Eclipse has a pretty huge following of developers actually working on Eclipse code, as such the sheer number of hands at play have built a rather impressive but bulky package. It isn't uncommon for Eclipse to take nearly a GB of disk space. Bulk though seems to be a theme for a lot of Java apps.

In any event I just wanted to point out that Java has brought us some interesting apps. It may be less than ideal but hardware has done much to make Java more interesting or maybe half respectable. Native code and apps still feel better in my experience.
post #62 of 88
Quote:
Originally Posted by SolipsismX View Post

I vote for the BASE-2 and BASE-10 for kilo, mega-, gigi-, et al. be addressed first. Using both 2^10 and 10^3 for kilo- is absurd. BASE-10 was first, and by a large margin, so I think the IEC Binary Prefixes of kebi-, mebi-, gibi-, et al. be utilized... but it might be easier to convince Americans to use the metric system for something other than 2 Liters of soda.


Actually the IEC can go to hell. I've never understood this issue in the first place. Context is everything.

I work on systems that use a lot of octal notation, sometimes it does confuse me especially if I come off another system where it isn't used. But who's fault is it if I forget to make the mental context shift? Frankly this is just another IEC proposal that the standards people came up with to justify their jobs. Computer systems have always been binary so there is no problem at all with using kilo mega and what have you in this context. IEC Binary Prefixes are a solution to a problem that never existed.
post #63 of 88
Quote:
Originally Posted by wizard69 View Post

Actually the IEC can go to hell. I've never understood this issue in the first place. Context is everything.

I work on systems that use a lot of octal notation, sometimes it does confuse me especially if I come off another system where it isn't used. But who's fault is it if I forget to make the mental context shift? Frankly this is just another IEC proposal that the standards people came up with to justify their jobs. Computer systems have always been binary so there is no problem at all with using kilo mega and what have you in this context. IEC Binary Prefixes are a solution to a problem that never existed.

Imagine if what you call octal was also called hex or binary. When it comes to a numerical value it's highly important that the terms aren't easily interchangeable. Saying kilo = 1024 after kilo was equal to 1000 should have never happened in the first place.

"The real haunted empire?  It's the New York Times." ~SockRolid

"There is no rule that says the best phones must have the largest screen." ~RoundaboutNow

Reply

"The real haunted empire?  It's the New York Times." ~SockRolid

"There is no rule that says the best phones must have the largest screen." ~RoundaboutNow

Reply
post #64 of 88
Quote:
Originally Posted by SolipsismX View Post


The only app I use that I think is even Java-based is a Pearson-Vue app for practice exams. That looks alright. The UI doesn't appear to be created in Java. I wish it was as I assume that would make it easier to port to Mac OS.

I had thought that Wireshark uses Java since it's so incredibly ugly and clunky but Wikipedia says the UI is GIMP. I wish there was a native app for Mac OS X (or Windows) that can do what Wireshark can do.

 

WebObjects that Apple uses is also a Java app :) I believe there are far more Java apps running on server than on desktop machines.

 

Wireshark is native app in sense that it does not run on a virtual machine. The GUI toolkit that it uses, GTK+, is also native. However, it is not native as in natively comes with OS X.(i.e. Cocoa)

post #65 of 88
Quote:
Originally Posted by wizard69 View Post

Computer systems have always been binary so there is no problem at all with using kilo mega and what have you in this context.

 

Unfortunately not all "kilo" in computer systems is 1024. When we are talking about network bandwidth, kilo is still 1000. So 1 kbps is 1000 bps. (network is part of the computer system right?)

post #66 of 88
Ah, a million thanks Soli and Wizard. I have a perfect grasp on it now.1bugeye.gif
melior diabolus quem scies
Reply
melior diabolus quem scies
Reply
post #67 of 88
Quote:
Originally Posted by SolipsismX View Post

I vote for the BASE-2 and BASE-10 for kilo, mega-, gigi-, et al. be addressed first. Using both 2^10 and 10^3 for kilo- is absurd. BASE-10 was first, and by a large margin, so I think the IEC Binary Prefixes of kebi-, mebi-, gibi-, et al. be utilized... but it might be easier to convince Americans to use the metric system for something other than 2 Liters of soda.

Which is a tangential issue, and I honestly don't think this is a big deal. It was only an issue in regards to reported hard drive capacity vs. what was said on the box, and that issue seems to have died down with better reporting in the computer dialogue.
post #68 of 88
Quote:
Originally Posted by JeffDM View Post

Which is a tangential issue, and I honestly don't think this is a big deal. It was only an issue in regards to reported hard drive capacity vs. what was said on the box, and that issue seems to have died down with better reporting in the computer dialogue.

The issue hasn't died so much as the discrepancy has gotten so large that it's harder to be misunderstood in regards to HDD capacities but we still issue here nearly every week. It wasn't too long ago that an article claimed that Android OS plus Samsung apps on the S4 used xGB by subtracting the remanding capacity as noted by the OS (BASE-2) by the initial NAND capacity (BASE-10).

"The real haunted empire?  It's the New York Times." ~SockRolid

"There is no rule that says the best phones must have the largest screen." ~RoundaboutNow

Reply

"The real haunted empire?  It's the New York Times." ~SockRolid

"There is no rule that says the best phones must have the largest screen." ~RoundaboutNow

Reply
post #69 of 88

Samsung backed the wrong boat on this one. AMD is already eating Intel's lunch in this market space and will only expand with its most recent line up to compete against Haswell's low power alternatives.

Meanwhile, GlobalFoundries just turned up the heat:

http://www.techpowerup.com/184728/globalfoundries-accelerates-adoption-of-20nm-lpm-and-14nm-xm-finfet-processes.html
 

 

Quote:
GLOBALFOUNDRIES Accelerates Adoption of 20nm-LPM and 14nm-XM FinFET Processes

At next week's 50th Design Automation Conference (DAC) in Austin, Texas, GLOBALFOUNDRIES will unveil a comprehensive set of certified design flows to support its most advanced manufacturing processes. The flows, jointly developed with the leading EDA providers, offer robust support for implementing designs in the company's 20nm low power process and its leading-edge 14nm-XM FinFET process. Working closely with Cadence Design Systems, Mentor Graphics and Synopsys, GLOBALFOUNDRIES has developed the flows to address the most pressing design challenges, including support for analog/mixed signal (AMS) design, and advanced digital designs, both with demonstration of the impact of double patterning on the flow.

The GLOBALFOUNDRIES design flows work with its process design kits (PDKs) to provide real examples that demonstrate the entire flow. The user can download the design database, the PDK, detailed documentation and multi-vendor scripts to learn how to set up and use the GLOBALFOUNDRIES design flow. The flows use open source examples and provide the customer with working, executable and customizable flows.
"As the developer of the industry's first modular 14nm FinFET technology and one of the leaders at 20nm, we understand that enabling designs at these advanced process nodes requires innovative methodologies to address unprecedented challenges," said Andy Brotman, vice president of design infrastructure at GLOBALFOUNDRIES. "By working with a new level of collaboration with EDA partners, we can provide enhanced insight into our manufacturing processes in order to fully leverage the capabilities of 20nm and 14nm manufacturing. This provides our mutual customers with the most efficient, productive and risk-reduced approach to achieving working silicon."

Production Ready AMS flow from specification to verification
To address the unique requirements of analog/mixed signal (AMS) design at advanced processes, GLOBALFOUNDRIES has enhanced its design flows to provide production quality scripts and packaged methodologies. The new reference flow establishes a working flow from specification to physical verification that has been taped out to be verified on working silicon.

The AMS reference flow provides comprehensive double pattern design guidelines. It gives overview of decomposition flow for both block level and chip level. The flow also addresses decomposition for different design styles. Recommendations for color balancing, hierarchical decomposition, ECO changes are discussed. The flows also present decomposition impact on DRC run time and resulted database size.

Notably, the reference flow includes support for efficiency and productivity improvements in the Cadence Virtuoso environment specifically for designing in a double patterned process. The flow includes support for Virtuoso Advanced Node 12.1 and provides efficient access to the tool's productivity benefits for physical design with real-time, color-aware layout. Circuit designers can assign "same net" constraints in the schematic, and the layout designers can meet these requirements as they create the physical view. Additionally, layout designers can take advantage of Virtuoso tool support for local interconnect, and advanced layout dependent effect management.

The flow also features interoperability with Mentor's Calibre nmDRC, nmLVS, and extraction products which address multipatterning requirements for both double and triple patterning. In addition special settings for analog design; auto-stitching and when to use it; and fill and color balancing are described in detail.

The AMS flow provides detailed information on parasitic extraction and layout dependent effects, both of which introduce new challenges at 20nm and 14nm. For parasitic extraction, the flows are described in detail and customizable scripts and examples demonstrate OA and DSPF back annotation. In addition the flows illustrate methodologies to predict layout-dependent effects during schematic design and methods to include full models in post layout extraction. PEX flows for Synopsys StarRC extraction, Cadence QRC and Mentor CalibrexRC are supported.

These flows serve as references to validate the correctness of the accompanying PDK as well as the vendor tools setup.

Sign-off ready RTL2GDSII flows that address double patterning
GLOBALFOUNDRIES is also making available new flows that support a complete RTL-to-GDSII design methodology for targeting its 20nm and 14nm manufacturing processes. The company worked with EDA vendors to certify the flows in their respective environments and provide a platform for optimized, technology-aware methodologies that take full advantage of the performance, power and area benefits of the processes.

The result is a set of fully executable flows containing all the scripts and template files required to develop an efficient methodology. The flows serve as a reference to validate the correctness of the accompanying PDK as well as the vendor tool setup. In addition the flows offer access to other critical and useful information, such as methodology tutorial papers; guidelines and methodologies for decomposition of double patterned layouts; PEX/STA methodology recommendations and scripts; and design guidelines and margin recommendations.

A critical aspect of manufacturing at this level is the use of double patterning, an increasingly necessary technique in the lithographic process at advanced nodes. Double patterning extends the ability to use current optical lithography systems and the GLOBALFOUNDRIES flows provide comprehensive double pattern design guidelines. They address design for double patterning and the added flow steps for different design styles and scenarios.

This includes support for odd cycle checking, a new type of DRC rule that must be met to allow for legal decomposition of the metals into two colors. This check is detailed in the flow and guidelines are provided to make sure it is met.

Synopsys and GLOBALFOUNDRIES worked together to minimize the impact of changes associated with the 3-D nature of FinFET devices as compared to planar transistors. The two companies focused on making FinFET adoption transparent to the design team. The collaboration on Synopsys' RTL to GDSII flow includes 3-D parasitic extraction with the Synopsys StarRC tool, SPICE modeling with the Synopsys HSPICE product, routing rules development with the Synopsys IC Compiler tool and static timing analysis with the Synopsys PrimeTime tool.

Cadence contributed a complete RTL-GDSII flow, including physical synthesis, and planning and routing developed with the Encounter Digital Implementation (EDI) System foundation flow. The seamless implementation flow, using Cadence Encounter RTL Compiler and EDI System, supports double patterning and advanced 20- and 14-nm routing rules.

Mentor's Olympus-SoC place and route system is supported in the flow, providing support for new DRC, double patterning, and DFM rules. The Olympus-SoC router has its own native coloring engine along with verification and conflict resolution engines that detect and automatically fix double patterning violations. Expanded features include DP-aware pattern matching, coloring aware pin access, pre-coloring of critical nets, and DP aware placement. The Calibre InRoute product allows Olympus-SoC customers to natively invoke Calibre signoff engines during design for efficient and faster manufacturing closure.

Double patterning also impacts LVS and other DRC issues, and the flows provide methodology details to address these areas, including hierarchical decomposition to reduce data base explosion. Parasitic extraction methodologies and scripts are provided as well, offering ways to address double patterning-induced variations via DPT corners or with maskshift PEX features.
post #70 of 88
Quote:
Originally Posted by Dave MacLachlan View Post

You'll pry my Imperial Measurements from my cold, dead fingers! 1smile.gif

Well you'll have to pry my binary correctness from my 10 cold, dead hands. 1tongue.gif

"The real haunted empire?  It's the New York Times." ~SockRolid

"There is no rule that says the best phones must have the largest screen." ~RoundaboutNow

Reply

"The real haunted empire?  It's the New York Times." ~SockRolid

"There is no rule that says the best phones must have the largest screen." ~RoundaboutNow

Reply
post #71 of 88
A x86 Android tablet can use the same apps that a ARM based tablet can. Dalvik, like Java is CPU architecture independent, the Asus Fonepad which is Atom based has no problems playing games like Nova 3, GTA Vice City, Max Payne, ect.
When I looked up "Ninjas" in Thesaurus.com, it said "Ninja's can't be found" Well played Ninjas, well played.
Reply
When I looked up "Ninjas" in Thesaurus.com, it said "Ninja's can't be found" Well played Ninjas, well played.
Reply
post #72 of 88
Quote:
Originally Posted by Rogifan View Post

I see we've veered off to SamsungInsider again...

But but but ... I don't understand why Apple users don't want to hear about the competition! Only rational and open-minded people appreciate competition (and the rest of you are iSheep). /s

"Apple should pull the plug on the iPhone."

John C. Dvorak, 2007
Reply

"Apple should pull the plug on the iPhone."

John C. Dvorak, 2007
Reply
post #73 of 88
Quote:
Originally Posted by SolipsismX View Post

jragosta is correct.

Running MS Windows on a PPC Mac is emulation because Windows only had binaries for x86. Running MS Windows on an Intel Mac is virtualization because Windows understands the HW. That means it al had to be emulated so that the processor code understand what the OS was requesting. This made it very slow even on the fastest systems. Many Intel chips are even designed to allow a VM to take advantage of the HW with less overhead but the OS needs to understand x86 or x86_64 HW for this to work.

No, there isn't ARM emulation happening. The Dalvik vm runs as a native x86 binary. The binary translator converts ARM machine code to equivalent X86 code.
post #74 of 88
Quote:
Originally Posted by Relic View Post

A x86 Android tablet can use the same apps that a ARM based tablet can. Dalvik, like Java is CPU architecture independent, the Asus Fonepad which is Atom based has no problems playing games like Nova 3, GTA Vice City, Max Payne, ect.

This is also true for all apps built and compiled with Android NDK?

edit: It appears Google has worked on allowing fat binaries for 3rd-party developers for their NDK. Apparently APP_ABI := all will compile binaries for ARMv5, ARMv7, x86_32 & MIPS_32.

Quote:
Originally Posted by nht View Post

No, there isn't ARM emulation happening. The Dalvik vm runs as a native x86 binary. The binary translator converts ARM machine code to equivalent X86 code.

I didn't say it was happening with Java. I said he's correct that emulation is not virtualization.
Edited by SolipsismX - 5/31/13 at 3:57pm

"The real haunted empire?  It's the New York Times." ~SockRolid

"There is no rule that says the best phones must have the largest screen." ~RoundaboutNow

Reply

"The real haunted empire?  It's the New York Times." ~SockRolid

"There is no rule that says the best phones must have the largest screen." ~RoundaboutNow

Reply
post #75 of 88
Quote:
Originally Posted by Tallest Skil View Post

So what does this mean? Incompatibility with everything made before it?

It means they're sacrificing battery longevity. Want proof? Compare Surface Pro and Surface RT. Night & Day.  

post #76 of 88
Quote:
Originally Posted by Gatorguy View Post

. . . and?
Would it then be using emulation or virtualization? I'm way over my head when it comes to technical issues like that and so would defer to you.

Cool. I found my new signature.
"I'm way over my head when it comes to technical issues like this"
Gatorguy 5/31/13
Reply
"I'm way over my head when it comes to technical issues like this"
Gatorguy 5/31/13
Reply
post #77 of 88
Quote:
Originally Posted by nht View Post

No, there isn't ARM emulation happening. The Dalvik vm runs as a native x86 binary. The binary translator converts ARM machine code to equivalent X86 code.

But you still have inefficiencies in the translation - so the translated app will generally run slower.

Not to mention that there will be incompatibilities and some apps won't run (even Anand's report confirms this). As just one example, if the app is based on Flash, it's going to have trouble - since the latest versions of Flash won't run on Dalvik. And since we heard endlessly how critical Flash was, that's an important deficiency.
"I'm way over my head when it comes to technical issues like this"
Gatorguy 5/31/13
Reply
"I'm way over my head when it comes to technical issues like this"
Gatorguy 5/31/13
Reply
post #78 of 88
Quote:
Originally Posted by dreyfus2 View Post

 

My point (it might well be irrelevant, what do I know) is simply that I would consider it surprising if Samsung would release a mass market device (supposed to compete with the iPad, Amazon's and Samsung's own ARM devices) that would be incompatible with quite a few Android apps, especially since Android is still short of dedicated tablet apps. What is the actual benefit of a x86 CPU when running Android? A product needs at least one selling point. It won't be lighter, it won't be cooler, it can't be cheaper, and it definitely won't have more battery life... so, what is the point?

 

Quote:

Originally Posted by AppleFanPro View Post

It means they're sacrificing battery longevity. Want proof? Compare Surface Pro and Surface RT. Night & Day.  

 

 

For Samsung to switch to Intel's Atom lineup for a tablet, there has to be an extremely compelling reason that they would give up using their own SoCs.  My guess is that it will have better performance and longer battery life than the competition.  And who knows what's in store for future versions with Bay Trail later this year.

 

The comments about Intel's battery life in Atom products isn't accurate.  If you are comparing a Surface Pro with Surface RT, remember that Surface Pro has an Intel Core i5 chip, not Atom.  If you read about the latest Intel Medfield chips in smart phones, you see they deliver the same battery life (with better performance) than ARM-based phones (for example, see the comparisons with the Motorola Razr M with Qualcomm versus Razr i with Intel).

 

It's a good point though - there must be something really spectacular about the chips that would Samsung choose them over their own chips.  My guess is that Samsung compared their own roadmap with Intel's and made the decision to switch.

post #79 of 88
Quote:
Originally Posted by mdriftmeyer View Post

Samsung backed the wrong boat on this one. AMD is already eating Intel's lunch in this market space and will only expand with its most recent line up to compete against Haswell's low power alternatives.

Meanwhile, GlobalFoundries just turned up the heat:

http://www.techpowerup.com/184728/globalfoundries-accelerates-adoption-of-20nm-lpm-and-14nm-xm-finfet-processes.html
 

 

 

Really?  Where is AMD eating Intel's lunch in anything except gaming consoles?  And Global Foundries is far behind TSMC, much less Intel.  It's one thing to have a press release about future process technologies... it's another to be shipping them in hundreds of millions of units (like Intel has been doing with FinFET for over a year).

 

If you are a device maker, being able to receive chips 1) on time and 2) in mass quantity is as important as how good the chip is.

post #80 of 88
Quote:
Originally Posted by SolipsismX View Post

.
I didn't say it was happening with Java. I said he's correct that emulation is not virtualization.

That's not what he said. He claimed that the intel android was running in emulation and not in a vm. This is and was clearly false. At no point was he correct.

You had no idea what a VM was, no understanding that the Dalvik VM is like the Java VM and further confused the issue by reinforcing the confusion between virtual machine based platforms/languages (Java/JVM, Java/Dalvik, C#/CLR) with virtualization.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: iPad
  • Samsung chooses Intel CPU for new iPad-competing Android Galaxy Tab - report
AppleInsider › Forums › Mobile › iPad › Samsung chooses Intel CPU for new iPad-competing Android Galaxy Tab - report