Jump to content

32 or 64 bit OS?


Recommended Posts

that's why i go for 65 just to be a little bit more better

 

but for true digital manipulation a 69 process is best

Link to post
Share on other sites
  • Replies 43
  • Created
  • Last Reply

Top Posters In This Topic

  • Cipro

    9

  • miles-high

    5

  • MattFromGA

    4

  • Denis

    4

Top Posters In This Topic

Popular Posts

When I only need 2G of addressable memory I use 32bit because 32bit is typically more compact.  Programs need a little less space than they do when they run 64bit.  However my personal desktop is alwa

Go for 64bit.   32bit can only access up to 4GB ram.

Not to mention that one can only actually use about 3.3GB of that 4GB.

  • Admin (Retired)

but for true digital manipulation a 69 process is best

guess i'll have to update to 70 .. but i may make a stop over at 69

Link to post
Share on other sites
miles-high

128 bit CPU is 'round the corner... 256 has been announced... hope I can live to use them for my Bejeweled xx...

 

 

guess i'll have to update to 70 .. but i may make a stop over at 69

 

I have been telling my 18yo friends that the higher the number the better... but they tell me the lower bits can make more impact statements...

Edited by miles-high
Link to post
Share on other sites
  • Admin (Retired)

 

I have been telling my 18yo friends that the higher the number the better... but they tell me the lower bits can make more impact statements...

 

 

there is a lot to be said for old school gaming.. a lot of times now days you will see pretty poor games covered up with amazing graphics and such

Link to post
Share on other sites
MattFromGA

 

First, when a window is obscured it's contents are not cached by the OS, for Windows. When an obscured portion of a device context (represented by an HDC, handle to device context, in the API) is made visible the owning winproc will be asked to repaint the view. This may be accomplished by a variety of application specific means but in no way is bitmap caching required or advised for most applications. It would, for instance, be silly for Word to cache a bitmap when it can easily re-render the visible portion of the document. See WM_PAINT for more info.    Second, assuming we have 2 of those racy new 4K displays, our display pixel count would be around 16mp, or 16,000,000 32 bit words of memory, or 64 megabytes. If we are generous and figure the driver and OS writers are complete dolts, we could give them 3x that for caching and still be under 200mb, which is barely noticeable on anything with 2+ gb, not to mention the fact that the window bitmaps are not cached to start with (first paragraph).    Finally, the contents of windows ARE sometimes cached (during dragging, for instance) by the 2D accelerator code in the GDI, but that is GPU memory,   Extra memory will allow things from the file system and so on to be cached though, which is nice. 

 

Are you the member Peguin running around with a new user id?

 

There are plenty of sites talking about how more pixels on your screen equals more memory used by the OS.  That is all I was trying to talk about.  You're educational diatribe comes off as some type of ego trip to me.  Now might be a great time to talk about the good versus bad of operator overloading in C++.  I'm sure other members here will be deeply fascinated by your depth of knowledge in that area too.

Link to post
Share on other sites
miles-high

 

I just bought a new computer and because I chose 64x i was told i had to go with Office 10 and not Office 13

 

MS Surface Pro 64 bit OS (Win8 or 8.1) comes pre-installed with Office 2013...

Link to post
Share on other sites

not so sure

 

this sums it up http://en.wikipedia.org/wiki/Physical_Address_Extension

 

and note "The original releases of Windows XP and Windows XP SP1 used PAE mode to

allow RAM to extend beyond the 4 GB address limit. However, it led to

compatibility problems with 3rd party drivers which led Microsoft to

remove this capability in Windows XP Service Pack 2."

 

Imnot sure apps would (or more like should) need to be aware of whats going on here and thus Im not 100% conviced about your compile time statement; i would havt that an OS APi would be more the thing you need, as the app would ask the os to swap pages.

 

but anyways it over 30 years since worrying baout paging was an integral part of my day time job. i kind of assumed that memory pafing would be evrywhere but looks like i was wrong on that point.

 

but in th end, people who say that you never need more than 4gb on a 32 bit machine probably dont understand well how computers work

 

From your Wiki link:

 

"Design: x86 processor hardware-architecture is augmented with additional address lines used to select the additional memory, so physical address size increases from 32 bits to 36 bits. This increases the physical memory addressable by the system from 4 GB to 64 GB ...."

 

Is this not precisely what I said?

 

If you run 64 bit Windows, your apps can happily be 32 bit Windows apps, just like if they were running under a PAE 36 bit system, but they are limited to a 4gb address space, just as they would be in a 36 bit system. The OS has to be more than 32 bit to access the bigger address space, and drivers need to be more than 32 bit aware too. It doesn't really matter between 36 (PAE) and 64, but since 36 bit addressing is no longer supported it's not useful to recommend it as an option. 

 

Without hardware and a platform (OS) that supports greater than 32 bit addressing, it is impossible to address more than 2^32 addresses, and when each address is a byte, the hard upper limit becomes 4gb. If you're only allowed 0-9 and 2 digits, you can't count past 99, it's that simple. 

 

And yes, x86/x64 has/have page tables. 

 

 

 

128 bit CPU is 'round the corner... 256 has been announced... hope I can live to use them for my Bejeweled xx...

 

 

 

I have been telling my 18yo friends that the higher the number the better... but they tell me the lower bits can make more impact statements...

 

We are already using 256 bit CPUs, in a way, just not 256bit addresses. Many operations on big data will be vectorized into parallel SIMD instructions and executed much faster. As for more than 64 bit addressing, well never say never but 2^64 is a truly large number. For instance a 1000 element loop multiplying one set of 32 bit numbers by another set can (often) transformed into a 250 step loop that stuffs 4 of those values into a 128 bit register and does 4 multiplications in one step. 

 

 

 

Are you the member Peguin running around with a new user id?

 

There are plenty of sites talking about how more pixels on your screen equals more memory used by the OS.  That is all I was trying to talk about.  You're educational diatribe comes off as some type of ego trip to me.  Now might be a great time to talk about the good versus bad of operator overloading in C++.  I'm sure other members here will be deeply fascinated by your depth of knowledge in that area too.

 

No, I am not Peguin. 

 

 

C++ is a bit high level for the concepts here, it's closer to assembly I guess, although C++ does support pointers it has a memory model and doesn't really expose the hardware addressing scheme. 

 

With iPhones becoming 64 bit I really don't see much use in installing 32 bit OS on a platform that's 64 bit capable, honestly. 

  • Like 1
Link to post
Share on other sites

I realize how important this computer jargon is to you computer nerds. But, can you fellas refrain from attacking each other regarding whose processor is bigger?

 

super-computer-nerd.jpeg

 

Let's just work together here and stay on topic for everyone. Thanks. 

Edited by Paul
  • Like 1
Link to post
Share on other sites

I realize how important this computer jargon is to you computer nerds. But, can you fellas refrain from attacking each other on whose processor is bigger?

 

super-computer-nerd.jpeg

 

Let's just work together here and stay on topic for everyone. Thanks. 

I'll have you know that's an old picture, my new PC is much sexier. 

  • Like 1
Link to post
Share on other sites

So bottom line and all dick waving aside, the rule I use is "if the system can run 64 bit, it should BE running 64 bit".

 

Why?

 

Many reasons but basically it boils down to the fact that if your CPU can run in 64 bit mode it was probably integrated into a system that, even IF 3gb or less is currently installed, can support more. And yes, for Windows users, 3gb is a good upper limit to use for 32 bit code because of issues with devices getting memory mapped, kernel space, and so on. 

 

So if you can, do, if not, don't. Simple. 

 

And yes, 32 bit systems are still being produced. Microsoft just gave me one last summer. 

Link to post
Share on other sites

64.

 

It's more than 32, so that's gotta be good, right?

It's DOUBLE, way more than going to 11. 

Link to post
Share on other sites

 

can you fellas refrain from attacking each other regarding whose processor is bigger?

 

I have to say that as far as I am concerned, as someone that was first taught to program using punch cards, that the introduction of the 8080/z80 and 6501/6502 was the real revolution and, simplistically, all developments after that are 'icing on the cake'.

 

When you have stuffed around with the likes of cp/m, xenix and apple basic it is a sheer joy to switch on my AMD 64bit computer and integrated GPU with a measly 8Gb of memory and 3TB hard drive and know that I have an operating system, Win7 64 bit or Debian 64bit, and it all just works.

 

What a joy!

  • Like 1
Link to post
Share on other sites

I have to say that as far as I am concerned, as someone that was first taught to program using punch cards, that the introduction of the 8080/z80 and 6501/6502 was the real revolution and, simplistically, all developments after that are 'icing on the cake'.

 

When you have stuffed around with the likes of cp/m, xenix and apple basic it is a sheer joy to switch on my AMD 64bit computer and integrated GPU with a measly 8Gb of memory and 3TB hard drive and know that I have an operating system, Win7 64 bit or Debian 64bit, and it all just works.

 

What a joy!

My first assembler targeted the 6502 - talk about a small register file, and wheee, 4k of memory. 

Link to post
Share on other sites

 

My first assembler targeted the 6502 - talk about a small register file, and wheee, 4k of memory.

 

LOL.... 4k memory... thems were the days....

 

I was lavish with 64k and NewDos 80 on my TRS80-I clone that was sold in Australia as System 80... 4 double density floppy disk drives...

 

The first hard disk drive I bought was a bargain at AUD$2,400 for 30MB but I had moved to an IBM XT clone by then....

 

and I still think the commodore 64 was the greatest games machine ever invented...

  • Like 1
Link to post
Share on other sites
Guest
This topic is now closed to further replies.

×
×
  • Create New...

Important Information

By using this site, you agree to our Terms of Use, Privacy Policy and Guidelines. We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue..