• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Bioshock DX9 vs DX10 benchmarks

Nice article, always like the Firing squad ones :).

They show the differences nicely, DX10 seems a lot smoother to me and general better look after playing the DX9 then then DX10.
 
Shocking how low the 2900XT is on those benchmarks... hopefully the new driver revision from AMD solves that a bit.
 
Still - it's interesting that using DX10 doesn't seem to lower performance for any of the cards they test.
 
I can tell that test is somewhat wrong simply from the water (The dx10 portion is most likely correct but dx9 isnt), in dx9 you do not get dynamic water ripples as shown in the screenshots, instead you get a weird splash ripple effect (I double checked this by running Bioshock in XP instead of Vista), If jihad could post that vid you will understand what I mean. Also my 2900 performs well at max settings 1280x1024 dx10 16xAF averaging 69fps. In dx9 I average 108fps...so it seems even with the -dx9 command it doesnt completely get rid of dx10 features.

Edit: Found it http://video.google.co.uk/videoplay...=32&start=0&num=10&so=0&type=search&plindex=5
 
Last edited:
Its possible that Vista is the problem, I would have rather seen the results of XP DX9 vs Vista DX10 tbh, but still nice to see that my box will perform rather well :)
 
Uriel said:
Still - it's interesting that using DX10 doesn't seem to lower performance for any of the cards they test.
Click to expand...

They got something wrong then .... for me DX10 is 25% slower than DX9 in vista (perhaps they used the in game switch thats doesn't actually run it in dx9, not the command line which does).
Vista dx9 is just a couple of fps below XP, so no biggie there.
 
Load of crap you can tell the review is flawed It looks like they had been using Vista and for whatever reason DX9 mode hadn't enabled properly. Look at the numbers their almost identical

http://www.firingsquad.com/hardware/bioshock_directx10_performance/page5.asp

The DX10 numbers are right on the 2900 but the DX9 numbers are way out. Like I said It dosn't look like DX9 was enabled properly in their review..

Just for comparison

DX10 Results 2900 vs GTS

1680x1050

Everything Maxed

2900 :

Min Max Avg
24 111 45.623

Those are the results dosnt run as good as DX9

Just to compare my mate did the same test at the same resolution as me on a GTS same in game settings his results were

GTS :

Avg: 46.885 - Min: 30 - Max: 108


As for DX9 alone my 2900 got with the Hotfix drivers...

Min: 36 Avg : 77 Max : 142

I believe this is even higher in the full game as It seems to perform better
 
Get a room you two :o :rolleyes: .

There is no point in taking benchmarks from reviewers systems and thinking "this is what I'll get with the same card" as there are too many variables to achieve this.

Benchmarks from reviewers are just a rough estimate what you should get with near to the same power.

I thought you'd both know this by now :o (joke :p).
 
Last edited:
Yea but you can clearly see DX9 mode hasn't enabled properly in Vista hence why the numbers havn't changed. Very fishy indeed. I also understand a few people have had problems getting DX9 mode to work I think Fornowagain was saying somthing about it on MSN the other night..
 
Last edited:
Tom|Nbk said:
Yea but you can clearly see DX9 mode hasn't enabled properly in Vista hence why the numbers havn't changed. Very fishy indeed.
Click to expand...

Can you perform the necessary tests on your system to clarify it being done properly?.
 
Tom|Nbk said:
Yea but you can clearly see DX9 mode hasn't enabled properly in Vista hence why the numbers havn't changed. Very fishy indeed. I also understand a few people have had problems getting DX9 mode to work I think Fornowagain was saying somthing about it on MSN the other night..
Click to expand...

also, how can you say dx9 mode isnt working? you can CLEARLY see it in the picture comparison of the particles on the chair
 
ergonomics said:
also, how can you say dx9 mode isnt working? you can CLEARLY see it in the picture comparison of the particles on the chair
Click to expand...
If you have a Dx10 capable card even when Dx10 is disabled via the in-game settings it still uses some of it apparently.

Putting -dx9 in the command line for the shortcut is the only way of pure Dx9 I read, I can only use AA when using the command line method.
 
Tom|Nbk said:
Yea but you can clearly see DX9 mode hasn't enabled properly in Vista hence why the numbers havn't changed. Very fishy indeed. I also understand a few people have had problems getting DX9 mode to work I think Fornowagain was saying somthing about it on MSN the other night..
Click to expand...


theres next to no freaking differences between dx9 and dx10 modes, so why would you expect a massive performance difference in the first place, if any?

fact is previous games, that have been horrific under dx10, are all crap. lost planet, terrible, terrible, terrible, terrible port. call of juarez not great and not a native dx10 app by any means. COH, adding lots of extra particles to be determined by the cpu, in an already cpu limited game, shock horror, adding cpu load, in a cpu limited game, caused horrendous performance.

this is a gpu limited game, with virtually nothing extra done while in dx10 mode, name a single good reason why performance should be that different at all?
 
drunkenmaster said:
theres next to no freaking differences between dx9 and dx10 modes, so why would you expect a massive performance difference in the first place, if any??
Click to expand...

Don't ask me ffs. Ask the developers.

ergonomics said:
if the 2900 was scoring higher than the 8800 gts i bet you wouldnt be saying the test was done wrong
Click to expand...

Nope I would not. Im not the only one to point out the test seems wrong. It looks like DX9 hasnt been totally enabled.
 
You must log in or register to reply here.
Back
Top Bottom