Recommended Posts

Windows 8 won't support 128bit processors since there's no reason to make them in the next few decades.

It's taken us 50 years to use 4GB of RAM, we won't jump to 16.8 million terabytes of RAM in 3 years.

Maybe not, but you can bet that Microsoft will add support for the processors in Windows 8.

Seeing as most people don't even use 16GB or more of memory in a 64-bit system yet, they still have 64-bit support.

Yes I do know the Turn Windows Features On or Off Dialog, however, AMD is working on 128-bit processors (Microsoft's 128-bit indiscretion) and Windows 8 and/or 9 is supposed to support that architecture. So, by windows 10 we won't need a 32-bit compatibility layer anyway, so why include one.

alright,

on the note .

FYI 16bit is not included in 64bit windows because of hardware limitation not software

Thats just it, it's a prediction, just like all the other posts on here.

Fact is, we just don't know anything about Windows 8, apart from the release date being three years away!!!!!!

Actually 128-bit processors could be here as soon as 2011, especially if AMD's Bulldozer and Intel's Haswell Architecture turns out to be 128-bit.

Hasewell are projected at earliest 2012,last i checked

128bit would be rather useful for high-ended Servers/Supercomputers

but not to much beside that atm.

hecks, Microsoft has just only recently dropped 32bit server release (WS 2008R2)

...

Hasewell are projected at earliest 2012,last i checked

128bit would be rather useful for high-ended Servers/Supercomputers

but not to much beside that atm.

hecks, Microsoft has just only recently dropped 32bit server release (WS 2008R2)

In 20 years they might be useful for supercomputers, but at the moment nobody in the world is anywhere near the limit.

We're talking about somebody using up 13,107,200,000 GB of RAM, that is very unlikely.

Edit: Whoops, got the number wrong, it's actually 17,203,200,000 GB of RAM.

Hasewell are projected at earliest 2012,last i checked

128bit would be rather useful for high-ended Servers/Supercomputers

but not to much beside that atm.

hecks, Microsoft has just only recently dropped 32bit server release (WS 2008R2)

Usually when they drop something from a server release, the client version follows, hence the reason why I think that windows 8 will be distributed as 64-bit only.

Edited by neo158

Why do people only think about memory address space when talking about 64bit and 128bit? Being able to use more ram isn't the only advantage to have the ability to crunch through larget ammounts of data per cycle. You do understand that a true 128bit CPU running any type of HPC app would crush any 64bit CPU?

Performance isn't just about how much RAM you can access. 128bit CPU's would be a vast improvment over what IT uses today in it's massive datacenters and mainframes. Reducing the need for insane ammounts of CPU's and blade servers, cutting back on power usage and overall costs.

Saying we don't need 128bit because we'll never have the need to use more ram than what we can get with 64bit for the next 20+ years is missing the whole point.

Why do you think there's this big push to using GPGPU mroe for things other than games?

So I know what the whole 128bit CPU thing is about, it's about register length for some instructions.

SSE and such already use 128bit registers and operations, so you can add or multiply 2 64bit floats (4 32bit floats, etc.) in one go, these future registers are 256bits, allowing for storage of 2 128bit integers or floats in a single register.

So the CPU will still be 64bit, but it can operate internally on 256bit registers containing 2 128bit values (same kinda way a 32/64bit CPU can operate on an 128bit register containing 2 64bit values). And apparently there are plans for future 512bit and 1024bit registers (512bit allows for 4 128bit floats, etc.)

So I know what the whole 128bit CPU thing is about, it's about register length for some instructions.

SSE and such already use 128bit registers and operations, so you can add or multiply 2 64bit floats (4 32bit floats, etc.) in one go, these future registers are 256bits, allowing for storage of 2 128bit integers or floats in a single register.

So the CPU will still be 64bit, but it can operate internally on 256bit registers containing 2 128bit values (same kinda way a 32/64bit CPU can operate on an 128bit register containing 2 64bit values). And apparently there are plans for future 512bit and 1024bit registers (512bit allows for 4 128bit floats, etc.)

that kind of huge :huh:

So I know what the whole 128bit CPU thing is about, it's about register length for some instructions.

SSE and such already use 128bit registers and operations, so you can add or multiply 2 64bit floats (4 32bit floats, etc.) in one go, these future registers are 256bits, allowing for storage of 2 128bit integers or floats in a single register.

So the CPU will still be 64bit, but it can operate internally on 256bit registers containing 2 128bit values (same kinda way a 32/64bit CPU can operate on an 128bit register containing 2 64bit values). And apparently there are plans for future 512bit and 1024bit registers (512bit allows for 4 128bit floats, etc.)

Exactly, it's all about running apps faster not about being able to use xGB of ram.

that kind of huge :huh:

It's a wide data type, but due to how it's used it's not really that big (a 128bit data type can store 4 32bit integers (say, for a 2x2 matrix), 2D drawing engines use 3x3 matrixes)

Exactly, it's all about running apps faster not about being able to use xGB of ram.

Well sure, but the CPU will still be 64bit. We don't call old Pentium's 128Bit CPU's since they have a 128Bit data type.

FYI 16bit is not included in 64bit windows because of hardware limitation not software

AMD64 processors can run 16-bit code. The reason it isn't supported in 64-bit Windows is because there is so little demand that adding support for it costs Microsoft more than they get back. Simple business decision.

AMD64 processors can run 16-bit code. The reason it isn't supported in 64-bit Windows is because there is so little demand that adding support for it costs Microsoft more than they get back. Simple business decision.

How does adding support for something they've owned for decades cost them anything? It's a simple subsystem, I can see it being removed for security or better stability but cost? Nah.

How does adding support for something they've owned for decades cost them anything? It's a simple subsystem, I can see it being removed for security or better stability but cost? Nah.

I don't think you fully understand the amount of Q&A and bugtesting that needs to go into every piece of code of a operating system especially one as widely used as Widnows where the economy of one of the worlds largest companies rely on that product. even if it's old ancient code known to be stable. and it still needs to be modified to fit inside the new OS and interpreters and the systems to support the sub system needs to be built. which requires more code work and even more bug testign and Q&A for something that there's absolutely ZERO demand or need for.

Just because something is relatively cheap compared to stuff like a new graphics system, new audio system, new shell/task bar and any number of big visible changes in vista and 7, doesn't mean that they don't cut it for the costs.

If they where to keep everythign that was relatively minor and relatively cheap, windows would be REALLy bloated, especially in code.

How does adding support for something they've owned for decades cost them anything? It's a simple subsystem, I can see it being removed for security or better stability but cost? Nah.

Because it has to be modified to work on the new platform. You can't just drop the old files in and have it work. Also, it's not a "simple" piece of code (and not a subsystem, Windows only has half a subsystem). It wasn't removed, it was never added.

  • 4 months later...

Is there any fresh news about Win 8 yet? Some leaked screenshots, or something? It's too quiet, that's a little unusual for a Microsoft product. whistle.gif

Was exactly the same thing with Windows 7. No official screens from MS before PDC, even though builds leaked months earlier. We've already seen some alleged post-RTM builds but they're visually identical to 7.

  • 2 months later...
This topic is now closed to further replies.
  • Recently Browsing   0 members

    • No registered users viewing this page.
  • Posts

    • VidCoder 12.20 by Razvan Serea  VidCoder is a DVD/Blu-ray ripping and video transcoding application for Windows. It uses HandBrake as its encoding engine. Calling directly into the HandBrake library gives it a more rich UI than the official HandBrake Windows GUI. VidCoder can rip DVDs but does not defeat the CSS encryption found in most commercial DVDs. You’ll need the NET 8 Desktop Runtime. If you don’t have it, VidCoder will prompt you to download and install it. The Portable version is self-contained and does not require any .NET Runtime to be installed. You do not need to install HandBrake for VidCoder to work. Feature list: Multi-threaded MP4, MKV containers Completely integrated encoding pipeline: everything is in one process and no huge intermediate temporary files H.264, H.265, MPEG-4, MPEG-2, VP8, Theora video Hardware-accelerated encoding with AMD VCE, Nvidia NVENC and Intel QuickSync AAC, MP3, Vorbis, AC3, FLAC audio encoding and AAC/AC3/MP3/DTS/DTS-HD passthrough Target bitrate, size or quality for video 2-pass encoding Decomb, detelecine, deinterlace, rotate, reflect, chroma smooth, colorspace filters Powerful batch encoding with simultaneous encodes Customizable Pickers to automatically pick audio and subtitle tracks, destination, titles and more Instant source previews Creates small encoded preview clips Pause, resume encoding VidCoder 12.20 changes: Updated HandBrake core to 1.11.2. Download: VidCoder 12.20 | 47.0 MB (Open Source) Download: Portable VidCoder 12.19 | 89.3 MB Link: VidCoder Home Page | Github | Screenshot Get alerted to all of our Software updates on Twitter at @NeowinSoftware
    • Too soon, I'm still not over this death!
    • Normally, I admit when a title is clickbait (unfortunately, it's become somewhat necessary to compete against AI-dominated news sections today), but in this case, all supported versions is implied and doesn't need to be spelled out in the title. Of course, I'm covering a Patch Tuesday update bug that is only available to supported Windows SKUs. All our coverage relates to supported Windows software and SKUs only unless we expressly state that it's "unsupported", "unofficial", or "third-party". I'm sorry, but supported/official SKUs don't need to be spelled out as such in every Neowin headline.
    • ALL versions or ALL SUPPORTED versions? Neowin does it again.
  • Recent Achievements

    • Week One Done
      Jordan Smith earned a badge
      Week One Done
    • Reacting Well
      BizSAR earned a badge
      Reacting Well
    • First Post
      AndreaB earned a badge
      First Post
    • Week One Done
      Huge Trailer earned a badge
      Week One Done
    • Week One Done
      Classifyskilleducation earned a badge
      Week One Done
  • Popular Contributors

    1. 1
      +primortal
      595
    2. 2
      +Edouard
      185
    3. 3
      PsYcHoKiLLa
      77
    4. 4
      Michael Scrip
      73
    5. 5
      Steven P.
      66
  • Tell a friend

    Love Neowin? Tell a friend!