28nm is still the cheapest node in per transistor terms.
That's not really true anymore. 14nm is cheaper for Intel to manufacture than 22nm (but Intel is the only company thus far with a mature, cost effective node at 14nm.) Remember that all the problems Intel had with ramping 14nm to high volume every other silicon fab will also experience.
Really what this tells us is that if you look at Intel's past two nodes (22nm and 14nm) they both have had about a 2 year, 6 month development cycle instead of the 2 year cycle we are used to. I think this is more just Intel being open with their customers and dealing with the 2.5 year cadence instead of the 2 year cadence by back-porting some of the new features that were originally going to debut in cannon lake to skylake. I think we can probably consider kaby lake to be skylake 2.0, hopefully more than devil's canyon was (actual new micro-architectural and/or chipset changes not just some new thermal compound and higher MHz.)
Honestly I think we will all be happier to have kaby lake next year than another 2 year wait like haswell and skylake.
Remember when everyone made fun of Mozilla because it had everything including the about:kitchensink in it? Remember how Firefox was supposed to get rid of all that bloat and modernize the web browser? Guess Mozilla is back to bundling a ton of junk together in to one package.
Only this time its far worse, at least with Mozilla it was useful stuff like a web browser and an HTML editor. This time we get junk of dubious value like Firefox Hello and Pocket which would be much better kept as downloadable extensions. Of course, it is painfully obvious that the reason they are not separate extensions is because of the financial upside they get from bundling them. Same thing with the "sponsored tabs."
I guess they just view Firefox as a cash cow that they need to milk to keep funding non-browser projects like that POS called Firefox OS, the Mozilla Science Lab, and all those grants they have given out over the years.
Oh and most their paid programmers/QA staff make little more than minimum wage. Just because Mozilla is "cool" doesn't make it okay to pay vastly under market value for their employee's services. Unfortunate to see how much Mozilla has become poisoned with mission creep and lack of a clear direction to the point that they have to lower themselves to the level of Java and bundle sponsored junk.
I don't use the Universal App API. So I have to ask. How is it worse than the model used by the Android and iOS API? Why wouldn't it be adequate for an app like Skype.
For basic calling functionality yes you could definitely get by with an Universal app. But remember that they sell a bunch of USB Skype phones that plug in to your desktop and have a keypad for dialing numbers and sometimes a LCD screen for contacts and/or video calls. There is pretty much no way you are getting stuff like that working with a Universal app.
The limited APIs and strict sand-boxing on universal apps limits the amount of actually useful software you can write for it. "Universal" really means lowest common denominator between our phone and desktop os. If all you care about running on your computer is cut the rope and angry birds then its fine. If you want an actual full featured computer... not so much.
Core i7 3960X
ASUS P9X79 WS
64GB RAM
nVidia GeForce 980 GTX
480GB Intel 730 SSD (used for OS and applications)
LSI PCIe SAS/RAID Controller
2x2TB 7200RPM WD Black HDDs in RAID1 array (used for data storage)
Blu Ray Burner
I use it for gaming, photo editing, video editing, and programming. What takes my Broadwell-U laptop 15min to compile this thing builds in 4min (our build scripts are multi-threaded.)
I seem to remember that during the presentation they explicitly stated that would be releasing a Linux version of the runtime libraries for Swift. At least that should give you the basics for a console/text user interface.
I doubt Apple is going to be making any GUI binding other than for Cocoa. I also doubt that the bindings for Cocoa will be included in the open source packages. It will be interesting to see how accepting they will be of community contributions.
Honestly I think those guys are a bunch of hypocrites. They make a big deal about openness and evil binary blobs etc. But last I checked I don't see their board design schematics, layout files, CAD drawings for the chassis, etc. available anywhere under an open source license. They make a big deal about needing open source to foil NSA backdoors... what about NSA backdoors baked in to the board design? A covert NFC chip can violate privacy just as easily.
Moreover, even if you had all the source code for every single byte of code, how can you trust the binaries that were pre-installed by these guys? Should we really expect every end user to recompile everything and crack open the case and use a flash programmer to reflash everything (flash programmers are spendy by the way.) Also, even if it is open source we won't find all the security vulnerabilities in the code anyway (see heart bleed.)
Call me crazy, but I respect IHVs wanting to have the ability to patch hardware issues on devices that have already shipped. Remember the Pentium FDIV bug? Intel has had up-datable microcode ever since then for a reason. Having hardware be patchable like that creates binary blobs of out necessity. I guess I'm just too pragmatic or something.
Your forgetting the best part of the original iPhone: it had the Internet. Not the "mobile" Internet, or the kinda sorta looks like the Internet... the Internet. It had the real Internet.
It really was something special. Its sad really how Apple has started stagnanting now that Steve Jobs is gone. It was also sad when towards the end of his life he started to go a little insane. Now I'm a former iPhone user typing this on my Android phone.
Totally 100% agree. That said, making it GPL would pretty much mean that any improvements made by outside contributors would not be included in Xcode. So you would instantly create a fork where everyone on Mac uses a different Clang then everyone else. That's not helpful for the Clang project, the LLVM project, the devs using Clang, or Apple. Also LLVM was already BSD before Clang showed up so licensing a sub project under same license as the parent makes sense.
Now don't get me wrong, I really really appreciate all the hard work the GNU project and the FSF have contributed, but to imply that _only_ the GNU project and the Linux kernel deserve credit seems quite unfair to me.
If we call in GNU/Linux because GNU deserves credit then we should name it GNU/Apache/Xorg/KDE/SystemD/Samba/LibreOffice/Mozilla/Linux, because all those other projects are just as critical to creating the modern, functional operating system that we have today.
Or we could grow up and just call it Linux because its just a name after all.
My theory is RMS and all his buddies over at the GNU project are still butt hurt about Linux stealing the thunder from GNU Hurd (25 years after the fact!) If they really want to have their GNU OS, then just finish Hurd already build your GNU package.
It's amazing how childish RMS can be sometimes, look at how he reacted to the fact that Clang and LLVM are now technically superior to GCC. Wrote a whiny blog post about how he admits it hurts on a personal level and then in the same paragraph attacks Clang as not being open source enough because it is BSD licensed instead of GPL! Honestly I think deep down inside RMS would have preferred that Apple kept Clang closed source even though he would never say that publicly. Apple gives us something for free that they totally didn't have to give us so obviously we should bite their hand off because they licensed it in a way that would allow them to continue using it in Xcode.
There is a lot of things I really like about the open source movement, but self righteous crap and the cliquey project leaders definitely leave a bad taste in my month.
The root cause of this problem and in fact most of the problems I've seen in the work environment is that most managers seem to be completely incapable of understanding the concept of a happy medium.
Honestly I'm not a fan of the 5x10 cube. The walls and the monitor are so close that there is absolutely no way for you to take a vision break and focus on something 20 feet away without getting up and getting a coffee, so you don't do it as much as you should. I can't think of many things that could possibly be worse for your eyes. And often those walls are 20+ years old, have never been cleaned, super dusty and all of them smell like B.O. At the same time, yeah I don't like being on display and having everyone constantly looking over my shoulders either.
Right now my team has our aisle set up with half height walls facing the walkway and doors, the desks set up so that we are looking out into the aisle. The back wall is full length. That way people walking down the aisle see the back of our monitors, not the front. But at the same time, I can look down the entire length of the aisle, which not only lets me see my other co-workers, its gives me distant objects to focus on every now and then without even having to stand up. This by far is the best of both worlds... a happy medium go figure!
Problem is everything has to be black and white all the time at the office. Nothing can ever been in between with management types because that is harder to brag about on your annual review. Either we are centralizing and standardizing on one solution for the whole company (even when that solution is a really bad choice for some things) or we innovating and being disruptive and not working together at all so that way we can duplicate a bunch of work. My suggestion that we should work together when it makes sense but don't force it always seems to fall on deaf ears.
So Microsoft starts laying off 18,000 employees in several waves starting in July this year. One of the first groups that was hit hard by layoffs was QA (mostly contract workers so they are easy to let go.) Within that, the QA department responsible for testing OS security patches was hit the hardest...
So now we are having a bunch of problems with botched updates that weren't tested sufficiently, go figure!
C is great for low level stuff since it is capable of generating machine code that has zero dependencies. K&R even explicitly mentions "non hosted mode" with no libc and implementation defined entrypoint semantics. In fact, it is the only language in mainstream use today that has this feature (aside from assembly.)
For kernel, drivers, firmware, embedded, and RTOS its pretty much the only choice unless you want to code everything in straight assembly. Since my current job is firmware programming, I've actually like the C language now that I've been forced to do a lot of heavy coding in it instead of Python.
How, exactly, can Nvidia make games run poorly on other hardware? They don't write the games. Both AMD and Nvidia have extensive outreach programs to developers and make engineers available to game studios
That is true, but nVidia's outreach engineers have a history of checking code that regresses performance on competitor hardware. See what this Value developer has to say about "Vendor A":
Vendor A is also jokingly known as the "Graphics Mafia". Be very careful if a dev from Vendor A gets embedded into your team. These guys are serious business.
How can using certain benchmarks (as you suggest) make games run slower on other hardware?
Thats not what I'm suggesting. I am suggesting that nVidia has a history of being dishonest which thier performance benchmarks. The worst case by far is during the GeForce FX era when they were caught a driver that detected it 3DMark 2001 and then only rendered content that was visible to the camera instead of the whole frame to boost thier benchmark scores. That was a while ago and I've been unable to find the original story on it.
AMD fanboy much?
Not at all, my desktop currently has a GeForce 570 installed. When I bought it nVidia clearly held the performance crown. That said, I really don't like the unethical business practices and I think I might not buy them again.
Honestly nVidia's business practices have been so shady recently, they make what Intel was doing in 2006 look tame. They do everything they can to make as many games as possible run poorly on anything other than nVidia GPUs, including sending employees to game studios to help "optimize" their games. They bribe review sites and "suggest" certain benchmarks to use. Also, thier legal department is more aggressive than Apple's. They deserve to be taken down a notch.
Given the fairly lame update to the Mac Mini caused mainly by the lack of choices in Intel's mobile CPU offerings (and Apple's refusal to design and stock a separate motherboard just for quad core)
Why would you be faulting Intel for this?!?
Not only that, but your argument is based on factually incorrect information. If Apple designed the new Mac Mini using the FCBGA1364 socket (high end mobile Haswell) instead of the FCBGA1168 socket (often referred to as Haswell ULT) then they could offer 2 core and 4 core minis without any board change.
The truth is Apple's designers care more about form and power dispation than having a quad core mini. Consider that the 13 inch Macbook Pro uses FCBGA1168 and the 15 inch uses FCBGA1364, just a screen size change is enough to justify a different socket and different inventory on the Macbook line up.
In fact, motherboard layout differences between the FCBGA1364 socket and the FCBGA1023 socket (used on the previous SandyBridge Mac Mini) are minimal compared to the amount of design change needed to go from FCBGA1023 to FCBGA1168. Apple has access to Intel's roadmaps >1 year before they are public, they knew that Intel would not have quad core on FCBGA1168 and they knew it would be more work to change thier design to use FCBGA1168 but they did it anyway. It was a deliberate design decision.
I no longer have faith in ANY of the conglomerates offering products all over the board.
Any conglomerate? What about 3M? They make a ton of stuff across the board and I buy a lot of their products (Scotch Tape, Post-It Notes, Scotch Brights, Nexcare & ACE bandages...) and honestly the only product I can think of that they dropped that I used a lot was their floppy disks... and they didn't really drop their floppy disk line they just sold it off as a separate company (Imation) and I was able to keep buying those floppy disks until floppys were pretty much dead and I no longer had any need for them.
When a conglomerate is well managed it actually works great, the problem is a lot of tech companies have tunnel vision and don't know how to manage a conglomerate.
Not a single person I have talked to still running 8.0 was even aware of the upgrade. It's not like they made a conscious choice to stick with 8.0, they simply didn't bother to even find out.
Guess you don't actually run 8.0 anymore (or you are domain joined) because on my 8.0 system a pop-up asking me to upgrade to 8.1 shows up every 2 hours after an Windows update a couple months ago.
There is actually one language that I can think of used to be popular and significant that is actually now dead: PL/M
CP/M was written in PL/M (the OS that MS-DOS is based on.) Later versions of CP/M had most of the code rewritten in assembly for speed reasons. When Microsoft converted it from the 8080 to the 8086 for PCs after version 1.0 one of the things they focused on was replacing the remaining PL/M code with C code. It didn't take much time before MS-DOS was completely free of PL/M code.
Fast forward to today and there isn't a single modern PL/M compiler out there. Pretty incredible really considering that today all it takes is 1 guy deciding to spend about 6 months writing a LLVM frontend. The last one was PL/M-386, which dates to the 80's, everything newer than that focuses on converting PL/M code to C code. I would be surprised to hear about a single new software project being started today in PL/M, and I expect that the number of programmers actively writing PL/M code is a 2 digit number.
Amazing when you think about it that a language used to implement an OS which the world's most popular OS is descended from is dead now.
...That said Microsoft would have to get the clue that developers have zero interest in Metro/Modern/Whatever apps, the environment is so limited that porting a Win32 app is basically as much work as porting a Win32 app to Android (esp. with stuff like Xarmarin, Qt, and other great cross platform libraries available to help) and nobody wants to pay MS 30% of their revenue and limit their distribution channel so strictly.
Sorry Microsoft management, I know leveraging market position in your core product line to push yourself in to a new market is one of the oldest tricks in your book. In this case, its trying to use regular Windows to push developers in to building software that is compatible with WinPhone so you have the catalog of 3rd party software needed to make WinPhone successful. Thing is in order for it to work this time Windows on tablets would need to be the universally preferred tablet OS. 10 years ago legacy Win32 compatibility would have been all you needed to be the preferred tablet OS, but since you gave the competition 3-4 years to build up a nice back-catalog of touch friendly 3rd party software Windows is NOT the preferred tablet OS, Android and iOS are.
You have nobody to blame except yourselves for giving your competitors that much time (well, maybe your former now retired CEO.) At this point just take a page from your buddies over at Intel, they made it so installing any arbitrary.apk on a x86 Android device just works (even if it has ARM native code.) And look, consumers are buying x86 Android tablets without a second thought since everything just works, hell a lot of the time an x86 Android tablet isn't even labelled Intel vs. ARM its so seamless. Make it so you can install Android.apks on Win8/RT/Phone, that will give you access to the software catalog you need to break in to the market. It would be even better if you could work about a deal to get Google Play on Windows... but I doubt Google will want to "play" with you at all:) The preferred route of making everyone else bend and do things your way its pretty much a non-starter at this point because you waited so long.
This would be a much more interesting technology for Windows RT as it would make those devices actually semi useful if the Windows back catalog was available on them.
I think what will be interesting and compelling for Broadwell Desktop is the Iris Pro graphics on LGA parts (not just BGA mobile parts like Haswell.) Certainly it won't be capable of competing with high end cards but you can probably expect mid range discrete graphics performance built in to the CPU.
For your standard desktop tower gaming rig it doesn't matter much since you will be likely using discrete graphics there anyway, what excites me more is mid range discrete graphics performance without the added power consumption->heat->large GPU heat sink. Which means a NUC form factor system with mid range discrete graphics performance, which would be a pretty awesome steam box and/or general living room entertainment system.
Also if Haswell history is any lesson, the chips with Iris Pro graphics launch after the chips with the low end integrated graphics. This probably gives Broadwell desktop a few extra months of life with the period in between Skylake desktop launch but before Skylake desktop with Iris Pro.
The issue with US manufacturing isn't the absolute dollar values/volume of goods produced, it is the trend line. US manufacturing isn't growing, its largely flat, that is the problem. We are naturally producing more humans constantly, but you can't hire those new humans in to an industry that isn't growing at the same rate as the population.
Slashdot Mirror
28nm is still the cheapest node in per transistor terms.
That's not really true anymore. 14nm is cheaper for Intel to manufacture than 22nm (but Intel is the only company thus far with a mature, cost effective node at 14nm.) Remember that all the problems Intel had with ramping 14nm to high volume every other silicon fab will also experience.
Really what this tells us is that if you look at Intel's past two nodes (22nm and 14nm) they both have had about a 2 year, 6 month development cycle instead of the 2 year cycle we are used to. I think this is more just Intel being open with their customers and dealing with the 2.5 year cadence instead of the 2 year cadence by back-porting some of the new features that were originally going to debut in cannon lake to skylake. I think we can probably consider kaby lake to be skylake 2.0, hopefully more than devil's canyon was (actual new micro-architectural and/or chipset changes not just some new thermal compound and higher MHz.)
Honestly I think we will all be happier to have kaby lake next year than another 2 year wait like haswell and skylake.
Remember when everyone made fun of Mozilla because it had everything including the about:kitchensink in it? Remember how Firefox was supposed to get rid of all that bloat and modernize the web browser? Guess Mozilla is back to bundling a ton of junk together in to one package.
Only this time its far worse, at least with Mozilla it was useful stuff like a web browser and an HTML editor. This time we get junk of dubious value like Firefox Hello and Pocket which would be much better kept as downloadable extensions. Of course, it is painfully obvious that the reason they are not separate extensions is because of the financial upside they get from bundling them. Same thing with the "sponsored tabs."
I guess they just view Firefox as a cash cow that they need to milk to keep funding non-browser projects like that POS called Firefox OS, the Mozilla Science Lab, and all those grants they have given out over the years.
Oh and most their paid programmers/QA staff make little more than minimum wage. Just because Mozilla is "cool" doesn't make it okay to pay vastly under market value for their employee's services. Unfortunate to see how much Mozilla has become poisoned with mission creep and lack of a clear direction to the point that they have to lower themselves to the level of Java and bundle sponsored junk.
I don't use the Universal App API. So I have to ask. How is it worse than the model used by the Android and iOS API? Why wouldn't it be adequate for an app like Skype.
For basic calling functionality yes you could definitely get by with an Universal app. But remember that they sell a bunch of USB Skype phones that plug in to your desktop and have a keypad for dialing numbers and sometimes a LCD screen for contacts and/or video calls. There is pretty much no way you are getting stuff like that working with a Universal app.
The limited APIs and strict sand-boxing on universal apps limits the amount of actually useful software you can write for it. "Universal" really means lowest common denominator between our phone and desktop os. If all you care about running on your computer is cut the rope and angry birds then its fine. If you want an actual full featured computer... not so much.
Core i7 3960X ASUS P9X79 WS 64GB RAM nVidia GeForce 980 GTX 480GB Intel 730 SSD (used for OS and applications) LSI PCIe SAS/RAID Controller 2x2TB 7200RPM WD Black HDDs in RAID1 array (used for data storage) Blu Ray Burner I use it for gaming, photo editing, video editing, and programming. What takes my Broadwell-U laptop 15min to compile this thing builds in 4min (our build scripts are multi-threaded.)
I seem to remember that during the presentation they explicitly stated that would be releasing a Linux version of the runtime libraries for Swift. At least that should give you the basics for a console/text user interface.
I doubt Apple is going to be making any GUI binding other than for Cocoa. I also doubt that the bindings for Cocoa will be included in the open source packages. It will be interesting to see how accepting they will be of community contributions.
I suggest rallying around vendors like this: https://www.crowdsupply.com/pu...
Honestly I think those guys are a bunch of hypocrites. They make a big deal about openness and evil binary blobs etc. But last I checked I don't see their board design schematics, layout files, CAD drawings for the chassis, etc. available anywhere under an open source license. They make a big deal about needing open source to foil NSA backdoors... what about NSA backdoors baked in to the board design? A covert NFC chip can violate privacy just as easily.
Moreover, even if you had all the source code for every single byte of code, how can you trust the binaries that were pre-installed by these guys? Should we really expect every end user to recompile everything and crack open the case and use a flash programmer to reflash everything (flash programmers are spendy by the way.) Also, even if it is open source we won't find all the security vulnerabilities in the code anyway (see heart bleed.)
Call me crazy, but I respect IHVs wanting to have the ability to patch hardware issues on devices that have already shipped. Remember the Pentium FDIV bug? Intel has had up-datable microcode ever since then for a reason. Having hardware be patchable like that creates binary blobs of out necessity. I guess I'm just too pragmatic or something.
Your forgetting the best part of the original iPhone: it had the Internet. Not the "mobile" Internet, or the kinda sorta looks like the Internet... the Internet. It had the real Internet.
http://m.youtube.com/watch?v=37fMdoU8kyY
It really was something special. Its sad really how Apple has started stagnanting now that Steve Jobs is gone. It was also sad when towards the end of his life he started to go a little insane. Now I'm a former iPhone user typing this on my Android phone.
Totally 100% agree. That said, making it GPL would pretty much mean that any improvements made by outside contributors would not be included in Xcode. So you would instantly create a fork where everyone on Mac uses a different Clang then everyone else. That's not helpful for the Clang project, the LLVM project, the devs using Clang, or Apple. Also LLVM was already BSD before Clang showed up so licensing a sub project under same license as the parent makes sense.
Now don't get me wrong, I really really appreciate all the hard work the GNU project and the FSF have contributed, but to imply that _only_ the GNU project and the Linux kernel deserve credit seems quite unfair to me.
If we call in GNU/Linux because GNU deserves credit then we should name it GNU/Apache/Xorg/KDE/SystemD/Samba/LibreOffice/Mozilla/Linux, because all those other projects are just as critical to creating the modern, functional operating system that we have today.
Or we could grow up and just call it Linux because its just a name after all.
My theory is RMS and all his buddies over at the GNU project are still butt hurt about Linux stealing the thunder from GNU Hurd (25 years after the fact!) If they really want to have their GNU OS, then just finish Hurd already build your GNU package.
It's amazing how childish RMS can be sometimes, look at how he reacted to the fact that Clang and LLVM are now technically superior to GCC. Wrote a whiny blog post about how he admits it hurts on a personal level and then in the same paragraph attacks Clang as not being open source enough because it is BSD licensed instead of GPL! Honestly I think deep down inside RMS would have preferred that Apple kept Clang closed source even though he would never say that publicly. Apple gives us something for free that they totally didn't have to give us so obviously we should bite their hand off because they licensed it in a way that would allow them to continue using it in Xcode.
There is a lot of things I really like about the open source movement, but self righteous crap and the cliquey project leaders definitely leave a bad taste in my month.
The root cause of this problem and in fact most of the problems I've seen in the work environment is that most managers seem to be completely incapable of understanding the concept of a happy medium.
Honestly I'm not a fan of the 5x10 cube. The walls and the monitor are so close that there is absolutely no way for you to take a vision break and focus on something 20 feet away without getting up and getting a coffee, so you don't do it as much as you should. I can't think of many things that could possibly be worse for your eyes. And often those walls are 20+ years old, have never been cleaned, super dusty and all of them smell like B.O. At the same time, yeah I don't like being on display and having everyone constantly looking over my shoulders either.
Right now my team has our aisle set up with half height walls facing the walkway and doors, the desks set up so that we are looking out into the aisle. The back wall is full length. That way people walking down the aisle see the back of our monitors, not the front. But at the same time, I can look down the entire length of the aisle, which not only lets me see my other co-workers, its gives me distant objects to focus on every now and then without even having to stand up. This by far is the best of both worlds... a happy medium go figure!
Problem is everything has to be black and white all the time at the office. Nothing can ever been in between with management types because that is harder to brag about on your annual review. Either we are centralizing and standardizing on one solution for the whole company (even when that solution is a really bad choice for some things) or we innovating and being disruptive and not working together at all so that way we can duplicate a bunch of work. My suggestion that we should work together when it makes sense but don't force it always seems to fall on deaf ears.
So Microsoft starts laying off 18,000 employees in several waves starting in July this year. One of the first groups that was hit hard by layoffs was QA (mostly contract workers so they are easy to let go.) Within that, the QA department responsible for testing OS security patches was hit the hardest...
So now we are having a bunch of problems with botched updates that weren't tested sufficiently, go figure!
C is great for low level stuff since it is capable of generating machine code that has zero dependencies. K&R even explicitly mentions "non hosted mode" with no libc and implementation defined entrypoint semantics. In fact, it is the only language in mainstream use today that has this feature (aside from assembly.)
For kernel, drivers, firmware, embedded, and RTOS its pretty much the only choice unless you want to code everything in straight assembly. Since my current job is firmware programming, I've actually like the C language now that I've been forced to do a lot of heavy coding in it instead of Python.
Don't invest in and artificially scarce commodity.
You mean Bitcoin?
I know I'm going to be modded down, but it had to be said.
How, exactly, can Nvidia make games run poorly on other hardware? They don't write the games. Both AMD and Nvidia have extensive outreach programs to developers and make engineers available to game studios
That is true, but nVidia's outreach engineers have a history of checking code that regresses performance on competitor hardware. See what this Value developer has to say about "Vendor A": Vendor A is also jokingly known as the "Graphics Mafia". Be very careful if a dev from Vendor A gets embedded into your team. These guys are serious business.
How can using certain benchmarks (as you suggest) make games run slower on other hardware?
Thats not what I'm suggesting. I am suggesting that nVidia has a history of being dishonest which thier performance benchmarks. The worst case by far is during the GeForce FX era when they were caught a driver that detected it 3DMark 2001 and then only rendered content that was visible to the camera instead of the whole frame to boost thier benchmark scores. That was a while ago and I've been unable to find the original story on it.
AMD fanboy much?
Not at all, my desktop currently has a GeForce 570 installed. When I bought it nVidia clearly held the performance crown. That said, I really don't like the unethical business practices and I think I might not buy them again.
Honestly nVidia's business practices have been so shady recently, they make what Intel was doing in 2006 look tame. They do everything they can to make as many games as possible run poorly on anything other than nVidia GPUs, including sending employees to game studios to help "optimize" their games. They bribe review sites and "suggest" certain benchmarks to use. Also, thier legal department is more aggressive than Apple's. They deserve to be taken down a notch.
Given the fairly lame update to the Mac Mini caused mainly by the lack of choices in Intel's mobile CPU offerings (and Apple's refusal to design and stock a separate motherboard just for quad core)
Why would you be faulting Intel for this?!?
Not only that, but your argument is based on factually incorrect information. If Apple designed the new Mac Mini using the FCBGA1364 socket (high end mobile Haswell) instead of the FCBGA1168 socket (often referred to as Haswell ULT) then they could offer 2 core and 4 core minis without any board change.
The truth is Apple's designers care more about form and power dispation than having a quad core mini. Consider that the 13 inch Macbook Pro uses FCBGA1168 and the 15 inch uses FCBGA1364, just a screen size change is enough to justify a different socket and different inventory on the Macbook line up.
In fact, motherboard layout differences between the FCBGA1364 socket and the FCBGA1023 socket (used on the previous SandyBridge Mac Mini) are minimal compared to the amount of design change needed to go from FCBGA1023 to FCBGA1168. Apple has access to Intel's roadmaps >1 year before they are public, they knew that Intel would not have quad core on FCBGA1168 and they knew it would be more work to change thier design to use FCBGA1168 but they did it anyway. It was a deliberate design decision.
I no longer have faith in ANY of the conglomerates offering products all over the board.
Any conglomerate? What about 3M? They make a ton of stuff across the board and I buy a lot of their products (Scotch Tape, Post-It Notes, Scotch Brights, Nexcare & ACE bandages...) and honestly the only product I can think of that they dropped that I used a lot was their floppy disks... and they didn't really drop their floppy disk line they just sold it off as a separate company (Imation) and I was able to keep buying those floppy disks until floppys were pretty much dead and I no longer had any need for them.
When a conglomerate is well managed it actually works great, the problem is a lot of tech companies have tunnel vision and don't know how to manage a conglomerate.
Not a single person I have talked to still running 8.0 was even aware of the upgrade. It's not like they made a conscious choice to stick with 8.0, they simply didn't bother to even find out.
Guess you don't actually run 8.0 anymore (or you are domain joined) because on my 8.0 system a pop-up asking me to upgrade to 8.1 shows up every 2 hours after an Windows update a couple months ago.
I'm really waiting for an x86 phone that can be bought in the USA.
Looks like your wish is coming true on Oct. 24: Intel Processors to Power New Asus Smartphone Hybrid for AT&T - TheStreet
There is actually one language that I can think of used to be popular and significant that is actually now dead: PL/M
CP/M was written in PL/M (the OS that MS-DOS is based on.) Later versions of CP/M had most of the code rewritten in assembly for speed reasons. When Microsoft converted it from the 8080 to the 8086 for PCs after version 1.0 one of the things they focused on was replacing the remaining PL/M code with C code. It didn't take much time before MS-DOS was completely free of PL/M code.
Fast forward to today and there isn't a single modern PL/M compiler out there. Pretty incredible really considering that today all it takes is 1 guy deciding to spend about 6 months writing a LLVM frontend. The last one was PL/M-386, which dates to the 80's, everything newer than that focuses on converting PL/M code to C code. I would be surprised to hear about a single new software project being started today in PL/M, and I expect that the number of programmers actively writing PL/M code is a 2 digit number.
Amazing when you think about it that a language used to implement an OS which the world's most popular OS is descended from is dead now.
...That said Microsoft would have to get the clue that developers have zero interest in Metro/Modern/Whatever apps, the environment is so limited that porting a Win32 app is basically as much work as porting a Win32 app to Android (esp. with stuff like Xarmarin, Qt, and other great cross platform libraries available to help) and nobody wants to pay MS 30% of their revenue and limit their distribution channel so strictly.
Sorry Microsoft management, I know leveraging market position in your core product line to push yourself in to a new market is one of the oldest tricks in your book. In this case, its trying to use regular Windows to push developers in to building software that is compatible with WinPhone so you have the catalog of 3rd party software needed to make WinPhone successful. Thing is in order for it to work this time Windows on tablets would need to be the universally preferred tablet OS. 10 years ago legacy Win32 compatibility would have been all you needed to be the preferred tablet OS, but since you gave the competition 3-4 years to build up a nice back-catalog of touch friendly 3rd party software Windows is NOT the preferred tablet OS, Android and iOS are.
You have nobody to blame except yourselves for giving your competitors that much time (well, maybe your former now retired CEO.) At this point just take a page from your buddies over at Intel, they made it so installing any arbitrary .apk on a x86 Android device just works (even if it has ARM native code.) And look, consumers are buying x86 Android tablets without a second thought since everything just works, hell a lot of the time an x86 Android tablet isn't even labelled Intel vs. ARM its so seamless. Make it so you can install Android .apks on Win8/RT/Phone, that will give you access to the software catalog you need to break in to the market. It would be even better if you could work about a deal to get Google Play on Windows... but I doubt Google will want to "play" with you at all :) The preferred route of making everyone else bend and do things your way its pretty much a non-starter at this point because you waited so long.
This would be a much more interesting technology for Windows RT as it would make those devices actually semi useful if the Windows back catalog was available on them.
I think what will be interesting and compelling for Broadwell Desktop is the Iris Pro graphics on LGA parts (not just BGA mobile parts like Haswell.) Certainly it won't be capable of competing with high end cards but you can probably expect mid range discrete graphics performance built in to the CPU.
For your standard desktop tower gaming rig it doesn't matter much since you will be likely using discrete graphics there anyway, what excites me more is mid range discrete graphics performance without the added power consumption->heat->large GPU heat sink. Which means a NUC form factor system with mid range discrete graphics performance, which would be a pretty awesome steam box and/or general living room entertainment system.
Also if Haswell history is any lesson, the chips with Iris Pro graphics launch after the chips with the low end integrated graphics. This probably gives Broadwell desktop a few extra months of life with the period in between Skylake desktop launch but before Skylake desktop with Iris Pro.
Here is a proper citiation: http://www.nationmaster.com/country-info/stats/Industry/Manufacturing-output
The issue with US manufacturing isn't the absolute dollar values/volume of goods produced, it is the trend line. US manufacturing isn't growing, its largely flat, that is the problem. We are naturally producing more humans constantly, but you can't hire those new humans in to an industry that isn't growing at the same rate as the population.