XP v Vista & 32 v 64

Hello FatRakoon, great stuff and I am sure everyone will be looking forward to seeing the results.

Just one thing I will mention though is, when running games that are capable of DirectX 10, make sure you run them in DirectX 9 as well as DirectX 10 when using Windows Vista. Since when comparing two different Operating System platforms with one another, the test conditions need to be as similar as possible i.e. the exact same settings, resolution and DirectX architecture to get accurate and conclusive results. :)

(Edit) Sorry I have just seen that you will be using a DirectX 9 card anyway. If anyone else is going to be doing any tests, please take the above into account anyway. :p

It may also be worth using each of the operating systems for a few days before doing tests. This is because; an operating system that has just been freshly installed may run perfectly fine and probably will for a few days. However, it may then start to degrade over time which is an extremely important thing to take into account since this will normally be the general performance the user will experience. A good operating system platform will retain its integrity 6, 12, 18 months after it has been freshly installed and in my opinion, that is what makes a good operating system. You also have to take into account that Windows Vista takes a few days to fully optimize itself and to learn the user's patterns. :)
 
Last edited:
Willy waving?
Im sorry, but as soon as I read that one, I lost it...
Ok, calm down now fatty, calm down...


Right then, where am I?

Yes, the specs of the PC I decided to use a DX9 card for 2 reasons...
1 - With XP, while there may very well be hacks or even that DX10 may have a future for XP, it does not at this time, have a place on it, so DX9 is the only fair way to do these tests at this time.
2-I dont have a DX10 card... LOL

Letting the PC settle in.
This is very true yes, however, what I have done, is a fresh install for all of them, so they all have their Nappy-rash so to speak. Vista has gone through its disk thrashing stage and each O/S has been left on over night ( Hence me needing to catch up with the Vista64 results ) before running any tests.
I will be doing updates on some of the tests for sure as I have done a couple as testers during the setups.


Points system? - Good call. I will update that shortly then.

Daughter has a friend over, got tea to sort out and then take her home, and picking a mate up so updates will be a couple of hours...
 
v0n said:
Well - it's a valid question - it's always just the few kilo, and look - we all pretty much need 2 to 4Gb of memory to feel comfy. If all your progs on x64 were on average 20 or 25% bigger, surely it's worth weighting the impact, no?
Click to expand...

I don't think anyone has ever said that x64 has the same memory requirements as x32. In fact I can remember saying here in the Windows forum when XP x64 came out that memory requirements are higher because doubling the word size is inevitably going to require more memory...

With memory as cheap as it is (4GB for 100 quid or so isn't it?) this really can't be considered as a "disadvantage" either.

20-30% is about the figure I would put on it too.
 
lowrider007 said:
Great stuff, I can't wait until you start filling in those gamer benchmarks :).
Click to expand...

Got a couple done.

Obviously, I can only do the games I own, plus with games like HalfLife etc that take a couple of centuries to install ta the best of times, and me having to have 4 installations at the same time on them... Its a bit pooey.

Im using the hocbench utils on the games where the Game itself has no benchmark. I can upload a couple more in a minute.
 
Bum.

Just gone to Bench FarCry.... I went into XP32, installed it. XP64 and installed it there, and then both Vista 32and 64.. Couple of hours later, they are ready and fully installed so Iwent to bench it and I need Patch 1.4 ? - Ok, another 130MB download and the 64Bits are not supported??? - OFFS!

Wasted time there!

Given that FarCry was was one the first to boast 64Bit support, and especially in that they showed off on just how good 64Bit FarCry is compared to 32Bit blah blah... That their own patch does not support 64Bit?

Swine.

Ok, next game...
 
BF2 Added to list

Source engine games are now in progress, however, not the quickest things to do, what with the updating etc...
 
FatRakoon said:
BF2 Added to list

Source engine games are now in progress, however, not the quickest things to do, what with the updating etc...
Click to expand...

I backed up my counter strike source install using steam :D. I'm about to do the FPS stress test once steam finishes installing.
 
I've spent some of this weekend compiling these benchmarks, not done that many tests only the basic ones. I threw Vista SP1 into the mix with 32-bit, I would do 64-bit but it was not working for me. Hopefully these are sort of comparable to fat rakoons. I would do more tests but I need my main rig back for the week :).

Heres the results of the benchmarks http://www.multi-game.co.uk/root/josh/bechmarks/benchmarks.htm
 
Hey cheers dude...

Yeah, I just spent all weekend trying to work through the code that filled up the webpage there ( LOL only joking - I gave up and just CTRL C/V'd the data ).

I still have a bunch of results I need to also update the page with too, but already I see a little pattern emerging that your setup has shown...

With 3DMark03 we see x64 edging out in front, and then x86 beating it by 3DM06

Vista32's SP1 certainly seems to have helped improve speeds a little over non-sp Vista ... especially in CS:S.

I have not still had the chance to try any Vista on the Source Engine... I have mostly been concentrating on the Doom engine games as I find that these are shining with x64.
Im still annoyed that the BenchMark Apps I am using, dont want to know the source engine games and yet they are among the best performers in 64Bit ( although on 2 of my PCs this is NOT the case )

Anyway, I have added your results on the page... For the time being, I have simply created a new table but I think that it might be a good idea to merge all the scores ... What say you?
 
FatRakoon said:
With 3DMark03 we see x64 edging out in front, and then x86 beating it by 3DM06

Vista32's SP1 certainly seems to have helped improve speeds a little over non-sp Vista ... especially in CS:S.
Click to expand...

From the 3DMarks data that we can see on both pages it looks like more samples will have to be taken for any conclusion at all - the two sets of 3D Mark 2003 have completely different outcome - according to your run XP64 is the fastest, Vista 64 is the slowest, according to 1337z0r's run Vista 64 is the fastest, while both XPs are the slowest. I'll try and see if I can get two identical machines for my runs, just to verify what kind of margin of error are we talking about in this particular benchmark...

Interesting that 1337z0r figures suggest SP1 is crippling Vista 32bit speed somewhat as well....

Anyway, I have added your results on the page... For the time being, I have simply created a new table but I think that it might be a good idea to merge all the scores ... What say you?
Click to expand...

We say - upload those pages already so we can see :)
 
FatRakoon said:
Anyway, I have added your results on the page... For the time being, I have simply created a new table but I think that it might be a good idea to merge all the scores ... What say you?
Click to expand...

Yep, sounds like a good idea to merge all the scores. If you have MSN add me it will be easier to sort out on there :).
 
Merge, as in merge into one number? I think it should be kept separate table considering the outcome is so different...
 
v0n said:
Merge, as in merge into one number? I think it should be kept separate table considering the outcome is so different...
Click to expand...

I mean that we store all the results in one place so they can be accessed and viewed easier, well thats what I mean.
 
ARGH!!!

The Mobo has died on me.

I just cannot get it to do anything ... I am going to have to swap Mobos over to another one and continue tests on another box.

I got a few results to post up too, so I need to get at those first.

I bloody knew I should have used another box cos this board has done nothing but give me headaches since day 1
 
You must log in or register to reply here.
Back
Top Bottom