32 bit is no longer valid

[apoppin]

Originally posted by: Avalon
What exactly will you guys be testing? I've got tons of benches and such I could possibly throw into the mix if you want.

nullpointerus and i are going to be using our rigs to test 4GB RAM gaming performance in Vista 32-bit vs. Vista 64-bit.

we will be setting up Vista 32- and 64- on separate partitions and comparing performance where [hopefully] the only variable is the OS. We have about 10 gaming benchmarks/games/demos to use between us.

i have been stuck on dial-up since i moved back to my rural home in SoCal, 8 years ago. Tonight, for the first time, i am attempting to pair my cell-phone via a USB Bluetooth adapter with my desktop so i can use my "unlimited data" plan and my cell-phone as a dial-up modem for it.
i actually got my bluetooth adapter just today. It's a 100 meter USB Belkin from a B&M CC for $40 - i asked the salesman expecting nothing, but he said he noticed something new recently came in. It claims to do exactly what i want. Unfortunately it wouldn't install in Vista, so i am doing the 4 hour D/L to get the new drivers.

i also got my new SATA burner and Nero 7 but am waiting to install after i [hopefully] get a faster connection to update them ... i *also* got both 64-Vista DVDs today but again, waiting to see if i can pull off a broadband connection for really cheap - or at all.


so i got lotsa work ahead ... but if i pull off broadband this way, my time spent d/l'ing Vista 64 updates will be cut by 90%

so my fingers are crossed

 

[Spike]

Originally posted by: apoppin

...
so i got lotsa work ahead ... but if i pull off broadband this way, my time spent d/l'ing Vista 64 updates will be cut by 90%

...

Just don't get screwed like this guy

 

[ViRGE]

Originally posted by: nullpointerus
Apple just released 64-bit iTunes:
http://www.apple.com/itunes/download/

But you must use IE to view that link due to Apple's stupid script that only spits out the 64-bit download link when the web browser reports itself as running on 64-bit Windows. A big...
:thumbsdown:
...to Apple for that little problem.

iTunes 7.6 still has NO support for 64-bit XP, unfortunately.

Hot diggity damn, I can finally use my Touch in my native Windows environment instead of needing VMWare.

I wouldn't expect or worry about XP64 support however, as has been said countless times before it's an orphaned OS. It's not even XP, it's Server 2003. For an end-user 64bit OS, you should be using Vista.

 

[Mem]

iTunes 7.6 still has NO support for 64-bit XP, unfortunately.

Not surprising since more companies are moving over to Vista x64 which is getting better support.

 

[Dadofamunky]

Originally posted by: zephyrprime

Originally posted by: Dacalo
I am going to jump to x64 as soon as MS ships my copy in a few months, but for now, 32 bit Vista will do. Have 512MB video card with 4GB RAM and Vista says I have around 3.25GB free. Not sure where 250 MB went.

That happens even with only a 256MB video card. Nvidia is wrong. The amount of virtual memory sucked up is always ~800MB as far as I can tell if your video card has more 256mb of memory or more. I've moved to 64bit now and I don't have these probs anymore.

I spoke to a colleague who worked at NVidia for five years as an engineer.

He states that no video card is ever going to occupy more than a few megs of RAM in system memory. No matter how big the frame buffer is. The system kernel does do a bunch of MALLOCs when the system is initialized to store critical routines for further use; after those routines are loaded the memory spaces for them are locked and those resources are managed in the normal flow of system operation. This will happen in any PC no matter what OS is used - 32 or 64-bit. Which reinforces our friend's argument that 64-bit is the way to go. You just have a lot more headroom.

But having two 512 MB or 768 MB video cards eating up the same amount of address space in system memory? It flat out does not happen. Ever. No video card ever uses more than a few megs.

 

[Dadofamunky]

And by the way, so far I ***:heart::heart::heart:*** Vista 64. I'm keeping my WinXP system for quite awhile longer; it runs great. But for me, Vista 64 is the future and it runs spectacularly compared to XP in the similar time period after release.

 

[Dadofamunky]

Originally posted by: taltamir

Originally posted by: Dadofamunky

Originally posted by: Pabster

Originally posted by: taltamir
iTunes cannot burn audio CDs on vista... because it uses its own driver for burning CDs, and thats only available in 32bits. It warns you that it is not compatible with 64bit OS but aside from not burning CDs it works perfectly... personally I wouldn't touch that piece of crap, but my brother who uses it on vista64 never had any trouble with it, ever (he doesnt need it to burn audio cds, not only does he never do so, he has nero in case he wanted to for some reason).

Yes, iTunes can burn Audio CDs on Vista, both 32 and 64 bit flavors. I know, because I do it all the time on x64 and I've done it on x86 several times too. The 64-bit requires the 64-bit GEAR drivers, and then it works just fine.

Pabster, that is a great tip. Thanks for posting that.

Read my reply too... I know about that 3rd party driver and I tried it... while it works it has the issue of casing crashes whenever you try to update itunes. So be aware that when you update itunes you have to uninstall it first.

Also a good tip, Taltamir.

Originally posted by: nullpointerus
Apple just released 64-bit iTunes:
http://www.apple.com/itunes/download/

But you must use IE to view that link due to Apple's stupid script that only spits out the 64-bit download link when the web browser reports itself as running on 64-bit Windows. A big...
:thumbsdown:
...to Apple for that little problem.

iTunes 7.6 still has NO support for 64-bit XP, unfortunately.

Yes! Yes! Yes! Woo hoo!

 

[Dadofamunky]

Sorry for double post, I'm suffering from a bad connection...

 

[wordsworm]

But having two 512 MB or 768 MB video cards eating up the same amount of address space in system memory? It flat out does not happen. Ever. No video card ever uses more than a few megs.

That's not true at all. I've looked at the address space and 320 MB in address space is allocated to my video card in four different sections with the greatest being 256. If you want, you can look for yourself.

How else would the computer deal with information if there was no address space? That's like having millions of unaddressed envelopes. The thing is that there is a maximum amount of addresses that can be had in 32 bit of 4.2 GB. That space is divided by the various components of a computer.

And I agree with Dadofamunky, Vista 64 is a fantastic OS compared to anything else I've tried from MS. However, I don't like the way user rights are diminished between XP and Vista. XP retail was pretty much yours for any one computer you ever had. Vista retail is good for any one machine for a maximum of 2. That really sucks in my books. Also, as far as I know, apoppin and JustaGeek are breaking the EULA and risk prosecution should they decide to publish their results unless they follow MS's apparently strict guidelines. However, from what I've read, this isn't new - as they've gone after independent labs in the past.

I really don't like what the lawyers have done with Vista. The software engineers, on the other hand, have really outdone themselves this time.

 

[apoppin]

Originally posted by: wordsworm
Also, as far as I know, apoppin and JustaGeek are breaking the EULA and risk prosecution should they decide to publish their results unless they follow MS's apparently strict guidelines. However, from what I've read, this isn't new - as they've gone after independent labs in the past.

we - Justageek is not involved [do you even bother to read the replies here? - you sure don't understand them] - are following MS' strict guidelines

i have an *upgrade* path from Vista 32 > Vista64. MS offers a 30-day trial. i am using that trial to publish my results and *will pay to upgrade* IF Vista 64 offers better gaming performance in my rig then Vista 32.

You would like us not to test it. My0f'nB
:roll:

 

[Dadofamunky]

Originally posted by: wordsworm

But having two 512 MB or 768 MB video cards eating up the same amount of address space in system memory? It flat out does not happen. Ever. No video card ever uses more than a few megs.

That's not true at all. I've looked at the address space and 320 MB in address space is allocated to my video card in four different sections with the greatest being 256. If you want, you can look for yourself.

How else would the computer deal with information if there was no address space? That's like having millions of unaddressed envelopes. The thing is that there is a maximum amount of addresses that can be had in 32 bit of 4.2 GB. That space is divided by the various components of a computer.

And I agree with Dadofamunky, Vista 64 is a fantastic OS compared to anything else I've tried from MS. However, I don't like the way user rights are diminished between XP and Vista. XP retail was pretty much yours for any one computer you ever had. Vista retail is good for any one machine for a maximum of 2. That really sucks in my books. Also, as far as I know, apoppin and JustaGeek are breaking the EULA and risk prosecution should they decide to publish their results unless they follow MS's apparently strict guidelines. However, from what I've read, this isn't new - as they've gone after independent labs in the past.

I really don't like what the lawyers have done with Vista. The software engineers, on the other hand, have really outdone themselves this time.

I'd probably agree with you on the first point if I hadn't asked a real live video card engineer about this issue. I tend to believe him.

Otherwise, I completely agree with you. However, I'm warming to Microsoft's UAC system in Vista jsut because there is a compelling reason for it. It forces application developers to write programs that can be installed and run without a Root account. (The Windows equivalent.) I don't have time to go into this, but I'm in favor of stronger security. The UAC is a good step forward, though it needs more work in the convenience department.

 

[JustaGeek]

Originally posted by: wordsworm

..................................................

And I agree with Dadofamunky, Vista 64 is a fantastic OS compared to anything else I've tried from MS. However, I don't like the way user rights are diminished between XP and Vista. XP retail was pretty much yours for any one computer you ever had. Vista retail is good for any one machine for a maximum of 2. That really sucks in my books. Also, as far as I know, apoppin and JustaGeek are breaking the EULA and risk prosecution should they decide to publish their results unless they follow MS's apparently strict guidelines. However, from what I've read, this isn't new - as they've gone after independent labs in the past.

I really don't like what the lawyers have done with Vista. The software engineers, on the other hand, have really outdone themselves this time.

:roll:

I wish you would stop embarrasing yourself, wordsworm...

Here is the very first comparison of performance, on my perfectly legal dual boot machine loaded with Windows Genuine Advantage approved Operating Systems in dual-boot configuration.

3DMark06 results in "invalid" 32-bit Windows XP versus totally "high end" 64-bit Windows Vista Ultimate.




...........................Windows XP Home x86...................Windows Vista Ultimate x64


Main Test Results

3DMark Score...............11602 3DMarks....................................10783 3DMarks

SM 2.0 Score......................5467 Marks........................................5137 Marks

SM 3.0 Score......................5262 Marks........................................5060 Marks

CPU Score..........................2630 Marks........................................2303 Marks


Detailed Test Results


Graphics Tests

1 - Return to Proxycon..........44.849 FPS......................................42.704 FPS

2 - Firefly Forest...................46.265 FPS......................................42.917 FPS


CPU Tests

CPU1 - Red Valley....................0.83 FPS.......................................0.727 FPS

CPU2 - Red Valley..................1.334 FPS.......................................1.167 FPS


HDR Tests

1 - Canyon Flight (SM 3.0).......50.427 FPS...................................48.143 FPS

2 - Deep Freeze (SM 3.0)..........54.82 FPS...................................53.049 FPS



Now - don't get me wrong. I like Vista, and gaming seems smoother than in XP. But we like numbers, and they speek for themselves - even if minute in terms of difference.

And this is just the beginning...

 

[SilentRunning]

Rig posted by: JustaGeek

-------------------------
C2D E6600 @ 2.925GHz (325x9) 1.3V Stock HSF
P5N-E SLI NVidia 650i BIOS 0401
3GB (4 sticks) G.Skill HZ@800MHz 4-4-4-12-2T @2.013V
BFG 8800 GT OC 512
22" ACER X221W @ 1680x1050
Sound Blaster Audigy 4 SB0610
320GB Seagate 7200.10 (XP Home x86)
300GB Seagate 7200.9 (Vista Ultimate x64)
Antec TP Trio 650W
COOLER MASTER Centurion 5 CAC-T05-WW w/ Side 80mm Fan

Quick questions:

Do you really only have 3GB installed?

Would not that negate one benefit of Vista x64?

Also, don't you think using the perpendicular recording hard drive for you XP test might give it an unfair advantage?

 

[JustaGeek]

Originally posted by: SilentRunning

Rig posted by: JustaGeek

-------------------------
C2D E6600 @ 2.925GHz (325x9) 1.3V Stock HSF
P5N-E SLI NVidia 650i BIOS 0401
3GB (4 sticks) G.Skill HZ@800MHz 4-4-4-12-2T @2.013V
BFG 8800 GT OC 512
22" ACER X221W @ 1680x1050
Sound Blaster Audigy 4 SB0610
320GB Seagate 7200.10 (XP Home x86)
300GB Seagate 7200.9 (Vista Ultimate x64)
Antec TP Trio 650W
COOLER MASTER Centurion 5 CAC-T05-WW w/ Side 80mm Fan

Quick questions:

Do you really only have 3GB installed?

Would not that negate one benefit of Vista x64?

Also, don't you think using the perpendicular recording hard drive for you XP test might give it an unfair advantage?

Yes, I do have only 3GB of RAM - I've been running Vista 64 for the past 3 weeks only.

3GB was fine for XP, and Vista would benefit from more than 4GB. The ~3-4GB area is reserved for the PCI memory, and most of the Vista drivers are written for the 32-bit OS anyway. They run in a *32 mode.

Only a few selected applications benefit from ex. 6 or 8 GB of RAM, and none of which I use.

Now - the Hard Drive.

NVidia driver shows the transfer rate as identical for both drives - about 77MB/s sustained, and ~180MB burst. And I do not think that the "load" time is part of the PCMark06 tests.

And lastly - please treat my personal experience as informational only. The "full fledged" testing is in progress as we speak - right, apoppin and nullpointerus...?

 

[apoppin]

And lastly - please treat my personal experience as informational only. The "full fledged" testing is in progress as we speak - right, apoppin and nullpointerus...?

i can speak for both of us ... we are still having some nagging little issues with our rigs that we need to clear up frirst - nullpointerus is reporting issues with the latest Cats, even to the 'white square' and i still have a BIOS issue to resolve that i (likely) just nailed down tonight as the culprit that will allow me to fix my SATA burner install].

The best news for me is that i *can* get AT&T wireless Internet using my cell phone as DUN!!!!!!! - 'officially'
--the bad news is that i have to pay $59 a month for "unlimited" access.
... but at least i can D/L at ~150kpbs instead of 5 ... i spent an hour with AT&T tech setting it up [after i already set it up correctly - only minus their connect number and "password" :roll: ] - but i now know how to set up bluetooth dial-up networking between a PC and a cell phone. ... just weird to see my cell phone permanently listed in "my computer" as "other".

it's legal ... and i *won't* get a $85,000 bill from AT&T wireless

oh yeah, the *bad news* i can't actually connect tonight ... the AT&T tower appears to be down [high winds] ... so i will try in the AM

 

[lopri]

Interesting discussion here. Although I haven't read through this thread, I used to use 8GB up till like half a year ago with a Q6600. Although I sometimes saw unbelievable amount of memory usage from certain games (~3GB), my main usage of Q6600+8GB was for virtual machines. I agree with some of the posters here that the transition to 64-bit has been long over due. If I have to lay the blame, it's gotta be on MS. Not only their support for 64-bit has been lacking (well actually more than lacking) but also they kinda screwed 64-bit with Vista: 1. They don't ship 64-bit disc with 32-bit disc except the retail Ultimate ($399!). You have to 'call' them or some such nonsense. 2. Vista wasn't polished enough at launch so people were weary of jumping onto it.

Also, there is the issue of hardware. Unlike A64 CPUs which are built ground-up for 64-bit OS and contain robust memory controller on-die (except Phenom, of course), Intel CPUs rely on MCH for memory management. Unfortunately, the MCH quality is severely lacking compared to their CPU, not to mention A64's IMC, and it's be very hard to get 8GB of RAM to work on current Intel desktop chipsets. (especially if you're overclocking or using dividers)

I know I'm probably talking off-topic here but I think the time is right for 64-bit. More than anything because the RAM prices are still record low. It's unfortunate that Microsoft OS isn't as robust as it should have been, but at least the situation looks getting better. Now they just need to lower the price of the freaking OS - Talking about Ultimate version obviously. MS cunningly omitted critical features from both Home Premium (lacks backup and security) and Business (lacks media center function), and for a power user Ultimate is the only option, IMO

 

[Mem]

I agree with some of the posters here that the transition to 64-bit has been long over due. If I have to lay the blame, it's gotta be on MS. Not only their support for 64-bit has been lacking (well actually more than lacking) but also they kinda screwed 64-bit with Vista: 1. They don't ship 64-bit disc with 32-bit disc except the retail Ultimate ($399!). You have to 'call' them or some such nonsense. 2. Vista wasn't polished enough at launch so people were weary of jumping onto it.

I don't think you can blame Microsoft too much,remember XP 64 bit ?...Microsoft have been trying to push 64 bit for awhile,however companies and consumer interest/demand is not high enough at the moment,down the road this will change.

Microsoft can only do so much,drivers are really down to the company of the hardware in question.
Should Vista have been 64 bit only,I would have to say no at this time but would agree they should of supplied a 64 bit DVD in every box.
I think Vista x64 is a great OS,really stable and I have no issues with drivers or lack of drivers etc...

End of the day we just have to wait until 64 bit support becomes more popular.
I have even read Microsoft's next OS Vienna is going to be available in both 32 bit and 64 bit formats just like Vista.

 

[wordsworm]

Apoppin: Did you run XP Home on DX 10? I thought not. As for comparing two OSes, I'll dig around for it. However, about posting comparisons of Windows OSes, check this out:

http://www.theregister.co.uk/2006/10/29/microsoft_vista_eula_analysis/

Whether or not 'they're watching you' is another issue altogether

Seems there is a mystery over how many times Vista retail can be uninstalled and reinstalled on a new machine. Anyways, it's confusing me.

 

[wordsworm]

Originally posted by: SilentRunning

Rig posted by: JustaGeek

-------------------------
C2D E6600 @ 2.925GHz (325x9) 1.3V Stock HSF
P5N-E SLI NVidia 650i BIOS 0401
3GB (4 sticks) G.Skill HZ@800MHz 4-4-4-12-2T @2.013V
BFG 8800 GT OC 512
22" ACER X221W @ 1680x1050
Sound Blaster Audigy 4 SB0610
320GB Seagate 7200.10 (XP Home x86)
300GB Seagate 7200.9 (Vista Ultimate x64)
Antec TP Trio 650W
COOLER MASTER Centurion 5 CAC-T05-WW w/ Side 80mm Fan

Quick questions:

Do you really only have 3GB installed?

Would not that negate one benefit of Vista x64?

Also, don't you think using the perpendicular recording hard drive for you XP test might give it an unfair advantage?

I think that's a part of the deal. I think that they want XP to win, so they're cutting Vista 64's advantages.

 

[nullpointerus]

Originally posted by: JustaGeek

Originally posted by: wordsworm

..................................................

And I agree with Dadofamunky, Vista 64 is a fantastic OS compared to anything else I've tried from MS. However, I don't like the way user rights are diminished between XP and Vista. XP retail was pretty much yours for any one computer you ever had. Vista retail is good for any one machine for a maximum of 2. That really sucks in my books. Also, as far as I know, apoppin and JustaGeek are breaking the EULA and risk prosecution should they decide to publish their results unless they follow MS's apparently strict guidelines. However, from what I've read, this isn't new - as they've gone after independent labs in the past.

I really don't like what the lawyers have done with Vista. The software engineers, on the other hand, have really outdone themselves this time.

:roll:

I wish you would stop embarrasing yourself, wordsworm...

Here is the very first comparison of performance, on my perfectly legal dual boot machine loaded with Windows Genuine Advantage approved Operating Systems in dual-boot configuration.

3DMark06 results in "invalid" 32-bit Windows XP versus totally "high end" 64-bit Windows Vista Ultimate.




...........................Windows XP Home x86...................Windows Vista Ultimate x64


Main Test Results

3DMark Score...............11602 3DMarks....................................10783 3DMarks

SM 2.0 Score......................5467 Marks........................................5137 Marks

SM 3.0 Score......................5262 Marks........................................5060 Marks

CPU Score..........................2630 Marks........................................2303 Marks


Detailed Test Results


Graphics Tests

1 - Return to Proxycon..........44.849 FPS......................................42.704 FPS

2 - Firefly Forest...................46.265 FPS......................................42.917 FPS


CPU Tests

CPU1 - Red Valley....................0.83 FPS.......................................0.727 FPS

CPU2 - Red Valley..................1.334 FPS.......................................1.167 FPS


HDR Tests

1 - Canyon Flight (SM 3.0).......50.427 FPS...................................48.143 FPS

2 - Deep Freeze (SM 3.0)..........54.82 FPS...................................53.049 FPS



Now - don't get me wrong. I like Vista, and gaming seems smoother than in XP. But we like numbers, and they speek for themselves - even if minute in terms of difference.

And this is just the beginning...

Hmm...some observations:

- You are comparing XP 32-bit to Vista 64-bit, which does not isolate the differences in 32-bit v. 64-bit.

- nVidia has always had a significant performance gap between their 64-bit drivers and their 32-bit drivers.

Nonetheless, your numbers ARE interesting. apoppin and I both have ATI HD2900XT's, so it would appear that, at some point, we'll need numbers from an nVidia 8800-series card.

 

[nullpointerus]

Originally posted by: JustaGeek

Originally posted by: SilentRunning

Rig posted by: JustaGeek

-------------------------
C2D E6600 @ 2.925GHz (325x9) 1.3V Stock HSF
P5N-E SLI NVidia 650i BIOS 0401
3GB (4 sticks) G.Skill HZ@800MHz 4-4-4-12-2T @2.013V
BFG 8800 GT OC 512
22" ACER X221W @ 1680x1050
Sound Blaster Audigy 4 SB0610
320GB Seagate 7200.10 (XP Home x86)
300GB Seagate 7200.9 (Vista Ultimate x64)
Antec TP Trio 650W
COOLER MASTER Centurion 5 CAC-T05-WW w/ Side 80mm Fan

Quick questions:

Do you really only have 3GB installed?

Would not that negate one benefit of Vista x64?

...

Yes, I do have only 3GB of RAM - I've been running Vista 64 for the past 3 weeks only.

3GB was fine for XP, and Vista would benefit from more than 4GB. The ~3-4GB area is reserved for the PCI memory, and most of the Vista drivers are written for the 32-bit OS anyway. They run in a *32 mode.

Only a few selected applications benefit from ex. 6 or 8 GB of RAM, and none of which I use.

...

Speaking in general terms, the addresses of the 3-4 GB region of RAM will get remapped to the 4-5 GB region of address space, so 64-bit Vista will see and use the full 4 GB RAM while devices still get to access the 3-4 GB virtual address space as if they were running in 32-bit mode. So, with 4 GB RAM, Vista 32-bit will miss 1 GB (more or less) whereas Vista 64-bit will see and use all 4 GB.

 

[chizow]

Originally posted by: JustaGeek
Now - don't get me wrong. I like Vista, and gaming seems smoother than in XP. But we like numbers, and they speek for themselves - even if minute in terms of difference.

And this is just the beginning...

Thats a great example of how not to test the benefits of more RAM and 64-bit. What have you actually shown in that example? XP vs Vista performance in a low-res GPU intensive demo. The disparity in 3DMark scores between XP and Vista are well-known, if you wanted to show differences in RAM or 32 vs 64-bit you would've run the same OS in 32-bit and 64-bit versions (Vista 32 vs. Vista 64, XP vs XP 64) and physically added/pulled RAM. At the very least, you would've timed the runs and run them at least 2x. Using only FPS to quantify the benefits of RAM is clearly flawed though. Simply put, if you were going to build a new rig with the goal of maximizing FPS, would you invest $600 in 4GB of DDR3 and use the onboard graphics or would you spend $500 on a high-end video card and get 4GB of DDR2?

You've already acknowledged perceived benefits with going to 64-bit Vista with the same amount of RAM as XP 32, yet your test doesn't really quantify those benefits. The first step would be to see if the app/game actually benefits from additional RAM or address space, which is why performance monitoring tools are important. Then you can begin to look at areas where there is a difference in performance, which again, are seldom quantified in FPS. Load times (from a save game, between maps, zone transitions etc) are one area you would see a benefit. Another is game play smoothness (thrashing, rubber-banding, hitching etc), which is much harder to measure and quantify, but easily observed. Its really not all that different from loading from the HDD, just on a smaller scale. Most convincing way is to witness the differences first-hand, second best would probably be a FRAPs recording (incredibly storage intensive). You can also use performance monitoring tools here to get a better understanding of what's going on with your system's memory/storage subsystems and how they correlate to things you see on screen. Hard Disk activity, hard faults, page inputs/reads etc. should correspond to what you see on screen, and if you have a loud HDD, like a Raptor or SCSI drive, what you hear as well.

Anyways, again it boils down to understanding how more RAM actually improves performance and setting expectations accordingly. Understanding the benefits and how it improves your gaming experience isn't easily quantified in the almighty FPS, but those who do understand the benefit know that more RAM used equates to better performance. After that, it really is as simple as looking at a performance monitoring app to know a system that uses more RAM will provide a better gameplay experience than a system using less RAM (or incapable of using more RAM). To what degree varies by game, but the difference is there quantifiable in FPS or not.

 

[apoppin]

Simply put, if you were going to build a new rig with the goal of maximizing FPS, would you invest $600 in 4GB of DDR3 and use the onboard graphics or would you spend $500 on a high-end video card and get 4GB of DDR2?

What does that have to do with anything?

... but the difference is there quantifiable in FPS or not.

so it's a "feeling" after all.




of course there is some difference - but does it make a 'practical' difference with today's 32-bit games? and i DO want to quantify it
--We are gonna run the tests you mention probably also looking at the FPS in terms of hi/low/average ... as well as timing the load/save times
... but i don't care to pull or add any RAM ... this is a test of Vista 32- vs 64-bit Vista in 4GB gaming rigs.

Anything else you want us to look at? i expect to test all weekend [assuming i can get everything 64-bit d/led in time]

 

[chizow]

Originally posted by: apoppin

Simply put, if you were going to build a new rig with the goal of maximizing FPS, would you invest $600 in 4GB of DDR3 and use the onboard graphics or would you spend $500 on a high-end video card and get 4GB of DDR2?

What does that have to do with anything?

Its pretty self-explanatory, that people point to FPS charts in the few published 64-bit vs 32-bit or 2GB vs. 4GB review comparisons and exclaim: "No FPS difference! No benefit!". Not only are comparisons based on solely FPS flawed, any conclusions drawn from them are as well. It makes as much as sense as upgrading from DDR2 to DDR3 and then expecting a massive (or even quantifiable) increase in FPS.

... but the difference is there quantifiable in FPS or not.

so it's a "feeling" after all.

Uh no, again, falling into the trap of the almighty FPS bar graph. If you somehow find a means to develop a tool that can accurately quantify repeated/skipped frames, hitching, or rubber-banding I'd love to see it. OTOH, its easily observable in both real game play and in FRAPs videos even if the almighty FPS indicator doesn't flinch.

of course there is some difference - but does it make a 'practical' difference with today's 32-bit games? and i DO want to quantify it
--We are gonna run the tests you mention probably also looking at the FPS in terms of hi/low/average ... as well as timing the load/save times
... but i don't care to pull or add any RAM ... this is a test of Vista 32- vs 64-bit Vista in 4GB gaming rigs.

Anything else you want us to look at? i expect to test all weekend [assuming i can get everything 64-bit d/led in time]

If a transition or stutter that takes place over a split second reduces the quality of your gaming experience, its a practical benefit. Will it show up in an FPS test? I highly doubt it. Even if it did, it would only be a momentary drop in frames lost when averaged in with thousands of other frames. In cases of rubber banding/hitching, how would you accurately quantify say, frames 1000 to 1010 being identical, then lurching forward at frame 1011 to where it should have been? Given a 60 FPS baseline, that 10 frames is only 1/6th of a second but the resulting jerkiness will be much more than that to the eye.

The point of pulling RAM is to get an idea of what quantifiable differences you can easily test to establish a baseline benchmark and refine your testing. Adding RAM back to 4GB in a 32-bit and 64-bit environment would then allow you to see if those results scaled or resulted in the same benefits you saw with more/less RAM.

Edit: In terms of load/zone times that benefit will be obvious if the game does in fact use additional RAM, which is why performance monitors are important. There should be some benefit from additional RAM from SuperFetch as well, but unless you can manipulate or fully understand how SuperFetch's caching algorithms work, its benefits will be unclear in a test bench environment where hardware/software are constantly changing with constant reboots.

 

[apoppin]

Originally posted by: chizow

Originally posted by: apoppin

Simply put, if you were going to build a new rig with the goal of maximizing FPS, would you invest $600 in 4GB of DDR3 and use the onboard graphics or would you spend $500 on a high-end video card and get 4GB of DDR2?

What does that have to do with anything?

Its pretty self-explanatory, that people point to FPS charts in the few published 64-bit vs 32-bit or 2GB vs. 4GB review comparisons and exclaim: "No FPS difference! No benefit!". Not only are comparisons based on solely FPS flawed, any conclusions drawn from them are as well. It makes as much as sense as upgrading from DDR2 to DDR3 and then expecting a massive (or even quantifiable) increase in FPS.

i see ... you were stretching something to the ridiculous to make a point. And i will have to agree that FPS are a "measure" of only part of the picture. However, FPS graphs will show a LOT more ... so the 'min/max/av' is pretty useful in a summary of results.

... but the difference is there quantifiable in FPS or not.

so it's a "feeling" after all.

Uh no, again, falling into the trap of the almighty FPS bar graph. If you somehow find a means to develop a tool that can accurately quantify repeated/skipped frames, hitching, or rubber-banding I'd love to see it. OTOH, its easily observable in both real game play and in FRAPs videos even if the almighty FPS indicator doesn't flinch.

i can't quite agree ... i believe watching the FPS in FRAPS [for example] does indicate repeated/skipped frames, hitching, or rubber-banding. Not to mention, a honest and attentive reviewer will mention issues related to gaming "smoothness"

of course there is some difference - but does it make a 'practical' difference with today's 32-bit games? and i DO want to quantify it
--We are gonna run the tests you mention probably also looking at the FPS in terms of hi/low/average ... as well as timing the load/save times
... but i don't care to pull or add any RAM ... this is a test of Vista 32- vs 64-bit Vista in 4GB gaming rigs.

Anything else you want us to look at? i expect to test all weekend [assuming i can get everything 64-bit d/led in time]

If a transition or stutter that takes place over a split second reduces the quality of your gaming experience, its a practical benefit. Will it show up in an FPS test? I highly doubt it. Even if it did, it would only be a momentary drop in frames lost when averaged in with thousands of other frames. In cases of rubber banding/hitching, how would you accurately quantify say, frames 1000 to 1010 being identical, then lurching forward at frame 1011 to where it should have been? Given a 60 FPS baseline, that 10 frames is only 1/6th of a second but the resulting jerkiness will be much more than that to the eye.

Well - logically - if you can't *notice* it, then it does not make a practical difference. i can only give my hopefully forthcoming experience with the 2 OSes. As to uploading FRAPS videos, i doubt it. Perhaps the graphs.

The point of pulling RAM is to get an idea of what quantifiable differences you can easily test to establish a baseline benchmark and refine your testing. Adding RAM back to 4GB in a 32-bit and 64-bit environment would then allow you to see if those results scaled or resulted in the same benefits you saw with more/less RAM.

i can only do either 2 or 4GB of RAM as i don't want to lose my Dual-channel capabilities. i think we're all pretty agreed that 2GB is the minimum for Vista gaming; 4GB is overkill for 32bit systems but we want to test the practical differences between it and the 64bit system addressing the entire 4GB. Perhaps since you have 8GB of RAM you might consider testing the differences for us vs 4GB in your rig.

Edit: In terms of load/zone times that benefit will be obvious if the game does in fact use additional RAM, which is why performance monitors are important. There should be some benefit from additional RAM from SuperFetch as well, but unless you can manipulate or fully understand how SuperFetch's caching algorithms work, its benefits will be unclear in a test bench environment where hardware/software are constantly changing with constant reboots.

[/quote]

i already know Ready Boost makes Hg:L's gaming performance *worse* ... what else will we find out? i am pretty eager to begin. It is not a matter of "ego" although i'd prefer to be "right". If 64 bit offers me a practical advantage over 32-bit Vista with my current rig and games then i will be taking advantage of MS' upgrade offer. i'll let you know.

BtW, there are generally not a lot of reboots ... you cold boot into Vista and let it settle down [a good while - watch the HD activity] ... run your game benchmark 3 or 4 times and average them ... then you can reboot and repeat to your heart's content; i just find something to do in the meantime [which is usually trying to make sense of the data].

Also, there are several ways to handle it ... you can create a pure gaming environment with no programs installed and run your benchmarks and do the rebooting between games - probably much like the tests tat AnandTech reviews - using a image to make identical environments ,,, or else you can make it a "average user's rig" with AntiVirus, background programs running and internet connected. Whatever i do, it will be the same between OSes so the variables are kept to a minimum

i'm more inclined to test the "average user's rig"