Vex Star

ok i need to know once and for all.. i’m building a gaming rig.. should i go with vista 32-bit or vista 64-bit???
yes thats a good idea cuz im a pc nub.. i know the basics and what not but when it comes down to 32-bit and 64-bit (scratches head).. i just dont know

you can address more memory

enhanced security
64-bit hardware + 64-bit OS + 64-bit programs = ftw
64-bit programs are getting plenty
have you ever played Far Cry in 64-bit mode? its epic man,… epic

the only significant difference is that 64-bit allows you to use more than 4Gb of memory. for the average user there is no reason to do it. no game is going to use more than 4Gb of memory simply because people running 64-bit Windows on a system they play games on is incredibly rare. it is usually reserved for servers, computers where people are working with 64-bit apps and large data (video editing, very large raw images, etc) and home users who either don’t know any better or have an epeen complex.

ahhh i see.. makes perfect sense.. thank you

picard.jpg

he’s talking about old shit
this is a gaming rig right?
most current and upcoming games that fully supports 64bit would run significantly faster, smoother.
its also more stable. ever experienced 5000 units moving in supcom?
its not just mainly addressing more ram, it will do more complex calculations per clock cycle

do you need it? no
would you get any benefits? yes

you missed out dude.
cliffs:

64-bit is the future. Might as well accept it.

i’ve never come across anything that mentioned any benefit to running 64-bit other than some strange security stuff and more memory.

srsly?

browsers, word, excel,… no diff. what you said/believe applies to these apps.

gaming? far cry is such a good example. 32bit vs 64bit

note that this is already old
I have mixed opinions on this topic. If I could go back in time I would have bought 32 bit Vista, but it’s not worth the trouble for me to download it/purchase it at this point. Just be aware that at some point you might have to compromise. Yeah, photoshop and shit is going to run like glass. I’ve had an overall smooth experience with 64 bit vista, but I’m telling you that 32 bit on my laptop has been smoother. The only reason to get 64 bit Vista is if you’re wanting to run 4GB of RAM. Anything less and I’d say stick with 32 bit. As to why Microsoft puts all of the editions of Vista on one disk instead of putting just the 64 bit and 32 bit versions of the OS you purchased? I don’t fucking know, and it would be nice if they’d do ONE thing that’s convenient for consumers just once.

Quote:
AMD’s goals are quite admirable, but the fact of the matter is that none of the visual improvements enabled by the Far Cry patches had anything to do with AMD64 or EM64T. They are artificially limited to run on those platforms alone, but could work just as well on a 32-bit platform.

Date: May 10th, 2005

dude…c’mon
Why is the date relevant? Didn’t you cite the article to highlight the difference between 32-bit and 64-bit? Yet, the article says there is no noticeable performance difference, and the visual difference has nothing to do with anything 64-bit.

it is relevant because you have to factor in that its old hardware + xp64 shitty drivers vs current hw + optimized 64-bit support.

and to quote crytek:

point being,… its not just more memory addressing as trouphaz said.

But, that has more to do with video card performance. They were already getting 150fps on "very high" settings, so the game wasn’t much of a burden on the video hardware to begin with.

64-bits is primarily used to increase address space. 64-bit integer math obviously benefits from 64-bit hardware. But, other than that, I don’t think there is much more to performance that can be attributed to 64-bits.
64-bit OSes are almost exclusively for accessing additional memory. now, i could see games running better if they have access to more memory, but there is no reason to believe that a 64-bit OS is going to perform better on its own unless MS is doing something odd with their OS.

so, if you install a 64-bit OS on a system with under 4Gb of memory, there should be no noticeable difference in performance.

well, that’s how the technology works.
you really have to look at the whole picture: more memory is being addressed; more registers in the cpu working; programs are coded and optimized to take advantage of it; etc…. all translates to breaking the 32-bit architectural limitations.

i’ve been working on 64-bit OSes for many years as a UNIX admin and the move to 64-bit was always about memory. now, it does help applications run faster if they have a need for more memory, but i never saw an increase if it was just a regular app. for example, we had some databases that had the 32-bit binaries and ran fine. when they ran the same ones with 64-bit binaries there was no performance change. then, on other large databses where they would benefit with a large SGA and we had 16Gb of memory on the system, then obviously the database would run more betterer being able to hold more in physical memory.

64-bits is only relatively new to the x86 world. 64-bit versions of UNIX have been around for 10 years while 64-bit hardware in the UNIX world have been around for 12 years or more.
trouphaz, database apps rarely use floating-point math because everything is quantized. There are a lot of desktop apps (anything involving 3D graphics, audio/video recording/rendering, and artificial intelligence) that use floating-point math, and having 64-bit quad-precision decimals really improves the quality of those operations.

Why do you think it was such a big deal when Macintosh (the #1 A/V workstation) went 64-bit with the G5 processor? Because it does make a difference.

Memory is a 32 vs 64-bit issue, number of registers and code optimization is not because those implementation details would affect performance regardless of whether the CPU is 32 or 64-bit. Moving from 32 to 64-bits obviously changes the size of the data you work with, but it doesn’t really change the way you manipulate that data. So, there really isn’t a dramatic change to the "big picture" as you might think.

Floating point isn’t a 32 vs 64-bit issue either because CPUs have had floating point units, with at least 80-bit double precision registers, built into them since the days of the 486.

Read what matlab has to say about floating point benchmarks. In theory, you’d expect a difference. But, the reality is that FP is handled by the FPU, and it works the same regardless of whether the CPU is in 32 or 64-bit mode.
Except that the floating-point value gets truncated when it gets shoved into a 32-bit register. Granted, it also gets truncated when it gets shoved into a 64-bit register, but more precision = better quality either way.

Fuck Matlab. Read what DigiDesign has to say about 64-bit processing. They actually use it for something.

yeah, as i was reading i started thinking about that. oh well. i’ll bow out of the argument as my experience is probably no longer applicable. i still think 64-bit Windows is more about epeen though.
for the sake of another data point…. I was a mapper on a quake3 mod team. On large maps like ours, the map compiler’s vis and light passes can take many hours to complete, eat a lot of ram if highres lightmaps are desired, and are also cpu intensive. Thankfully the compiler is multithreaded and open source. Compiling the app as an x64 native binary decreased compile times by 20-25% on average using the same hardware. amd64 isn’t just about addressing space, it gives an additional 8 gp registers as well as other optimizations.

No related posts.

Related posts brought to you by Yet Another Related Posts Plugin.

February 22nd, 2009 | Tags: 3D graphics, AMD, artificial intelligence, G5 processor, Microsoft, Microsoft Vista, Microsoft Windows, ram, technology works, UNIX, video hardware | Category: general | Subscribe to comments | Leave a comment | Trackback URL