XP Pro 64 or Longhorn

[ViRGE]

Originally posted by: bsobel

But it isn't linear growth, it's exponential. The average RAM configuration doubled nearly every 12-18 months; which is why we burnt through 32bit space in 20 years. At that same rate, it would take 40 years to burn through 64bit space, so the only way we can burn through 64bit space any faster is to end up on a 8-12 month timeline, which I don't see happening. I actually think it's more likely that 64bit space will last longer than 40 years, but we'll see.

Meet you back here on 4/14/2025 to see where we are then 😉

I've been on these forums for nearly 10 years now, another 20 is nothing.😀 :beer:

 

[dawks]

Originally posted by: Link19

I second this, I think M$ needs to get some features out and all that. About the BLOAT, wouldn;t it be easier if we went only text based and the apps still can run in its current state? How's that for resource saving?

That's not what I'm talking about. What I mean is more things stuffed into the OS that run on system startup by default and a more GUI animated desktop. I like a nice lokking GUI and click and point interface, but what we have now is fancy enough, and doesn't need to get any more fancy to add to the bloat. What's the whole point of releasing XP 64 if there aren't any programs that are going to take advanatge of 64-bit access to memory for quite some time? Miswell just wait for Longhorn 64-bit if it is still going to be a while before applications take advanatge of 64-bit access to memory. That is why I can see Longhorn being bloated and containing unnecessary features. Why else would XP 64-bit be released today and made to have a bright long future if Longhonr is going to be the OS of the future for 64-bit computing?

You saying "the GUI we have now is fancy enough" is like that CEO, back in the mid 70's saying "no one will ever need/want a personal computer at home". Or that one infamous quote of the 19th century "everything that can be invented has already been invented". Current GUIs are incredibly simplistic, and not very intuitive. Progress is a good thing, even if some don't agree with the direction.

As its been said, the people that are going to be making 64bit applications need a 64bit OS to develop on. Just as the 64bit OS developers needed a 64bit processor to develop the OS on. And often times, a technology needs a giant to get behind it, to make it successful.

Originally posted by: Link19

To give developers a 64-bit Windows platform to develop on, you can't have 64-bit apps until you have a 64-bit system to run them on. But even so most apps will hopefully never be 64-bit because there's no reason for them to be, for instance what would Word need >4G VM addressing for? Things like databases, 3D renderers, etc that actually need the additional VM are already 64-bit on unix so very little will change with the release of XP64 other than many people crying that they can't find 64-bit Windows drivers. Eventually games will probably have 64-bit releases but I doubt that'll happen soon either because they'll still need 32-bit versions for compatibility purposes.

So, does that mean that there is absolutely no benefit to developing 64-bit applications unless you have more than 4GB of RAM in the system? Because 32-bit applictaions can access up to 4GB of RAM? So would a 64-bit application have no performance advanatge over a 32-bit application unless you have more than 4GB of RAM?

And yea, other then the ability to address more ram, there isnt a huge advantage over switching from 32bit to 64bit. Most people are mistaken when they think its going to change the world, and they must switch now.

Originally posted by: ViRGE

Originally posted by: bsobel

Well, when you put things in perspective on the x86 side, we jumped to 32bits in 1985(386), and didn't make another jump until nearly 20 years later. This time around we're adding twice as many bits, so it should in theory take us twice as long(40 years) to outgrow 64bit addressing.

That presumes the industry has grown linerally, it hasn't...
Bill

But it isn't linear growth, it's exponential. The average RAM configuration doubled nearly every 12-18 months; which is why we burnt through 32bit space in 20 years. At that same rate, it would take 40 years to burn through 64bit space, so the only way we can burn through 64bit space any faster is to end up on a 8-12 month timeline, which I don't see happening. I actually think it's more likely that 64bit space will last longer than 40 years, but we'll see.

Forgive me if I'm wrong, my memory on my A+ is a little fuzzy, but didnt the change from 16bit to 32bit also improve not only addressing, but memory space protection too? Like real memory and protected memory or something similar? That jump was more then just additional memory access ability.. Its one of the key reasons NT is much more stable then 95/98, since 95/98 were actually based on 16bit MS-DOS?

 

[ViRGE]

Originally posted by: dawks
Forgive me if I'm wrong, my memory on my A+ is a little fuzzy, but didnt the change from 16bit to 32bit also improve not only addressing, but memory space protection too? Like real memory and protected memory or something similar? That jump was more then just additional memory access ability.. Its one of the key reasons NT is much more stable then 95/98, since 95/98 were actually based on 16bit MS-DOS?

Sort of, protected mode was introduced with the 286, but since it couldn't switch back to real mode from protected mode, it wasn't used much on the 286, and MS decided to bundle the use protected mode with 32bit addressing(just get it all over with), resulting in 386-enhanced mode. Windows 3.0 and up used protected mode, NT just better utilized it by being built from the ground-up for it, not to mention better the better logical control of multi-tasking NT's thread scheduler was designed with.