>4GB ram in win x86 possible?


Recommended Posts

is unlikely considering that 64-bit processors have wider registers (64-bits) to push more data through the pipes. It should be the same or better.

The amount of data doesn't affect the speed at which a smaller amount of data goes through it. It's pretty close to the same. It does depend on the processor arch and there is a tiny translation penalty for 32 bit apps.

Of course, there's also the portion of memory the GPU takes, 32 bit apps with LAA and all that to consider so it's not a simple comparison.

you know its possible for Pentium Pro processor can have up to 64 GB memory,

but only 4 GB can be accessed at a time.

PAE is an ancient old technolgy (1995), its already there before x86-64 CPU exisited,

but Microsoft only actively used it in their 64-bit OS.

Its not a Software limitation

yes it is, unless you're talking about CPU before 1998 that is.

XP pre-dates multicore CPUs, it isn't optimised for them in the slightest (let alone low power/scaling stuff which CPUs do now). Newer versions of Windows make much better use of the underlying hardware (DWM offloads all desktop drawing to the GPU, Direct2D/DirectWrite offloads all application drawing to the GPU, etc.)

What a load of crap. Multiple CPU systems were in use for many many years, the difference to the OS between a different CPU and a different core is very small.

What a load of crap. Multiple CPU systems were in use for many many years, the difference to the OS between a different CPU and a different core is very small.

For server class systems, yes. But the desktop certainly didn't have multiple core/CPU systems as the norm.

When XP was released, a system with 4 CPU cores and 4GB+ of RAM was a high end server and well outside the design specifications.

  • Like 1

Slumdog - Normally people are willing to give some consideration to alternative views, but you have pretty much the entire of the posters against you on this.

32bit will run more slowly on x64 enabled chips. The x64 chips either have less registers for x86 instructions, or virtualise the calls and instructions. Since the Core range, this has been the case, but it's becoming more pronounced over time as Intel and AMD start to push for more x64 usage. I am aware there was a register issue with certain batches of Core processors, but the Core2 range had fixed that and begun this trend.

On top of this, XP doesn't have access to DirectDraw/Direct2D, thus it IS slower for more complex graphical browser based workloads. Chrome and Firefox both leverage varying levels of Direct2D or proprietary hardware acceleration. Thus, both will be faster on 7/8.

A lot of the people in this thread are extremely knowledgeable in a vast array of disciplines. Pretty much everyone is saying you're wrong on this. If you choose not to regard this point of view, fine, but you shouldn't expect us to blow smoke up your ass to validate your views.

PAE cannot (to the best of my knowledge) be enabled on XP. Consider 64 bit systems, and consider Windows 7. Even if you don't like the interface, there are significant improvements to thread handling, core allocation, memory management, graphical interfaces, Direct X draw and layers. Hell, consider 8 and user Start8 or whatever you want. I don't care about the interface disagreements, but the kernel is much improved.

Even if you insist on using FIrefox in CPU rendering modes, you will find noteworthy performance improvements from the better process scheduling on 7.

Lastly: Even if you enable PAE on XP, there are issues with process limitations (something people mistook for threads.. seriously) and most apps are compiled to only address 2GB of space at a max. Unless you enable particular compiler flags, most compilers will automatically compile for a maximum of 2gb of addressable space per process.

So.. Time to move on.

EDIT:: It's worth noting my CPU knowledge is a little out of date. The_decryptor pointed out some changes to the Core iX processors that invalidated much of what I said about performance overheads. Either way, x64 is still faster >.<

  • Like 2

For server class systems, yes. But the desktop certainly didn't have multiple core/CPU systems as the norm.

When XP was released, a system with 4 CPU cores and 4GB+ of RAM was a high end server and well outside the design specifications.

I remember when XP came out, I installed it on a 2 year old laptop. It had a 300mhz processor, and 64mb of RAM. At school, the best computer was a 900mhz PC with something insane like 128mb of RAM. A year later, I was talking to a friend who built a computer with 256mb of RAM in his computer. This was enough for him to disable his page file.

I mean, I guess it's like using puppy Linux as your main OS. I don't know why but to each their own.

I remember when XP came out, I installed it on a 2 year old laptop. It had a 300mhz processor, and 64mb of RAM. At school, the best computer was a 900mhz PC with something insane like 128mb of RAM. A year later, I was talking to a friend who built a computer with 256mb of RAM in his computer. This was enough for him to disable his page file.

I mean, I guess it's like using puppy Linux as your main OS. I don't know why but to each their own.

My first build had 1GB of RAM on a Prescott based Pentium 4. I was blown away by the speed, XP screamed along.

I disabled PageFile and was blown away with the speed :D Now days, You don't need to disable paging because your system will have so much memory it only pages when MS's algorithm is being ****..

I liked XP (hated it when I first started using it) but 7 was a mile forwards in all regards. Vista was an improvement, but Microsoft has their priorities wrong. Even if you don't like Vista, there's nothing not to like about 7 >.<

EDIT::

Might I just add Greenwizard, your app isn't really a productivity app, some might contend it does the exact opposite :p

The amount of data doesn't affect the speed at which a smaller amount of data goes through it. It's pretty close to the same. It does depend on the processor arch and there is a tiny translation penalty for 32 bit apps.

But, any 64-bit data can now fit in a register. Unmentioned, but 64-bit processors also tend to have more registers as well.

What?

Can't use more than 4GB RAM with a 32-bit OS. If you want to use more, you'll have to dump that dinosaur of an OS you seem hell bent on using.

You missed the part where he said x64.. You know.. Being capable of more than 4GB of addressable memory O.o

You missed the part where he said x64.. You know.. Being capable of more than 4GB of addressable memory O.o

No I didn't. I was surprised that anyone would say that XP x64 was a good OS. The comment below "What?" was to the OP. Will edit.

No I didn't. I was surprised that anyone would say that XP x64 was a good OS. The comment below "What?" was to the OP. Will edit.

Windows XP Professional x64 Edition (XP x64) isn't bad at all. Windows XP 64-bit Edition (XP 64-bit) on the other hand had isn't all that good because it wasn't supported well.

there are kernel hacks floating around, which enable PAE support in xp/vista/7 x86, but as others said, PAE has been disabled for a reason in the consumer editions of windows.

although the hacks do work, it's not advisable to use them, if you want to use the pc for multimedia/gaming stuff, because it's very likely you will run into incompatible drivers.

-andy-

It's almost like all his knowledge of computer and OS stopped at 2005. Welcome to 2013. Why are you hell bend on keeping Windows XP? What's SO good about XP that keeps you from upgrading to Win7 x64?

high standards and unwillingness to compromise... if i have to drop the standards might as well move to win8 x64. and yes im already triple booting these 3OSes so there is no question of being stuck in 2005.

btw where did you pull out the year 2005 from?

dated is still faster if you use a web browser other than ie.... which i suppose everyone does ALL the time.

im looking for hacks mainly.

Then read that article. It tells you how to enable it by editing your boot.ini file.

EDIT: Whoops, this one does: http://msdn.microsof...8(v=vs.85).aspx

2005 about is when the XP kernel and GDI was deprecated and Longhorn development was underway, I think.

Windows XP Professional x64 Edition (XP x64) isn't bad at all. Windows XP 64-bit Edition (XP 64-bit) on the other hand had isn't all that good because it wasn't supported well.

Well one of those is for Itanium processors, there's probably nobody who uses those aside in certain very, very specific server environments.

It's almost like all his knowledge of computer and OS stopped at 2005. Welcome to 2013. Why are you hell bend on keeping Windows XP? What's SO good about XP that keeps you from upgrading to Win7 x64?

He's one of the reasons why after 12 years, XP still has ~40% market share. The scary thing is that there are many, many systems admins that think like this, and mistake their own stubbornness, and unwillingness to adapt, as maintaining high standards.

He's one of the reasons why after 12 years, XP still has ~40% market share. The scary thing is that there are many, many systems admins that think like this, and mistake their own stubbornness, and unwillingness to adapt, as maintaining high standards.

scared for whom? srsly :shiftyninja:

More to the point, there are lots of people with not very much money >.<

Enabling PAE on XP will only enable access to 4GB of RAM >.>

pae is automatically enabled on xp sp3 + x86_64 processors... this thread is about going beyond 4gb... where is the vast array of expertise when after 5 pages the subject of the thread is not even grasped.? :/

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

    • No registered users viewing this page.
  • Posts

    • Qualcomm takes on NVIDIA with new Dragonfly CPU and AI chips by Pradeep Viswanathan Microsoft, Google, Amazon, AMD, Meta, Apple, OpenAI, and several others have been developing their own chips for AI infrastructure. However, NVIDIA still remains the dominant player in the market. Today, Qualcomm announced a major expansion of its data center infrastructure portfolio to better compete with NVIDIA. The new lineup includes the Qualcomm Dragonfly C1000 CPU, Qualcomm High Bandwidth Compute technology, the Dragonfly AI300 inference accelerator, new connectivity products, and custom silicon solutions. Qualcomm claims that this new lineup improves performance per watt, token throughput, and total cost of ownership for AI data centers. The Dragonfly C1000 is a new data center CPU built with Qualcomm’s custom Oryon cores. This chip will feature more than 250 cores, frequencies above 5GHz, and a chiplet-based design. Qualcomm claims that this new C1000 can deliver more than 2x better performance per watt compared to existing server CPU offerings based on specifications. The Dragonfly C1000 will support PCIe Gen 7 with more than 2TB/s of connectivity, along with CXL, advanced RAS features, and both air and liquid cooling. Qualcomm expects the Dragonfly C1000 to be commercially available in 2028. Additionally, Qualcomm and Meta announced a multi-year, multi-generation agreement under which Qualcomm will supply Dragonfly C1000 data center CPUs for Meta’s next-generation server fleet. Qualcomm also announced High Bandwidth Compute, a new near-memory computing architecture designed to address AI’s memory bandwidth bottleneck. HBC Gen 1 will debut with the Dragonfly AI250, which is expected to sample in mid-2027. The AI250 will deliver 133TB/s per card, an 18x increase in effective memory bandwidth compared to the AI200 with LPDDR5X. The new Dragonfly AI300 with HBC Gen 2 is a rack-level AI inference platform from Qualcomm. Qualcomm claims that the AI300 can deliver 4x to 8x better performance per watt compared to existing GPU-based architectures based on memory bandwidth per watt per card. The Dragonfly AI300 is expected to be available in 2028.
    • IBM reveals sub-1nm chip technology, production expected in another 5 years by Pradeep Viswanathan TSMC is now leading the chip manufacturing industry with its 2nm-class process node called N2. Samsung Foundry also has a 2nm-class process node called SF2. TSMC says N2 entered volume production in Q4 2025. Samsung says SF2 started mass production in 2025. Today, IBM announced the world’s first sub-1-nanometer chip technology, marking another major semiconductor research milestone. The new technology is based on a 0.7nm, or 7-angstrom, node and uses a new transistor architecture called “nanostack.” The new design vertically stacks and staggers nanosheet-based transistors so that more components can fit into the same chip area while also improving performance and power efficiency. IBM claims that this new sub-1nm chip can pack nearly 100 billion transistors onto a chip the size of a fingernail. This offers almost twice the density, up to 50 percent higher performance, or 70 percent better energy efficiency when compared to IBM's 2nm node design announced back in 2021. Also, IBM mentioned that this new architecture can deliver 40 percent SRAM scaling. It is important to consider that this announcement from IBM is a research milestone rather than a near-term process node launch. Back in 2021, IBM unveiled the world’s first 2nm chip design, claiming 50 billion transistors on a fingernail-sized chip and major performance and efficiency gains. Five years later, IBM’s 2nm technology has still not entered mainstream commercial production. That is because IBM is no longer a major commercial chip manufacturer. It sold its chip manufacturing business to GlobalFoundries years ago and has since then focused only on semiconductor research, IP development, and partnerships. To productize its 2-nm chip technology, IBM partnered with Japan’s Rapidus, but it has not resulted in anything shipping at scale. IBM says that its new sub-1nm technology can reach production as early as within the next five years. If that happens, it will likely depend on manufacturing partners, advanced EUV tooling, and years of yield improvements.
  • Recent Achievements

    • Week One Done
      Meta Plast earned a badge
      Week One Done
    • First Post
      kinowa earned a badge
      First Post
    • Rookie
      krychek57 went up a rank
      Rookie
    • Grand Master
      Jaybonaut went up a rank
      Grand Master
    • One Year In
      Philsl earned a badge
      One Year In
  • Popular Contributors

    1. 1
      +primortal
      454
    2. 2
      +Edouard
      170
    3. 3
      PsYcHoKiLLa
      135
    4. 4
      Michael Scrip
      78
    5. 5
      Xenon
      77
  • Tell a friend

    Love Neowin? Tell a friend!