Seen something about Conroe, kinda struck me as odd

[classy]

I am a hardware junkie so I read 30 or so websites a day. Now I have my questions about Conroe and when I came across this I raised my eyebrows. Most of the results we have seen involved sythetic benchmarks or dual gpu configurations. I was reading over at ocworkbench and came across an article where Intel had a conference showcasing their new architecture in Bangkok. The interesting part was for the first time that I have seen they showcased a machine with only one video card a single 7800gtx, seen here. Here's the kicker, they demoed the machine against a 950 and they tested the machine in Quake 4. At 1024x768 it got 133.7 frames a second at medium settings, see here. Now it doesn't say whether or not the game was configured for dual core. So I checked a few sites to see what other systems scored using single cards, I found this at xbitlabs, where a FX60 using 4xAA and 16xAF at 1024x768 got 115.4 using a 7800gtx. And I also found this at firingsquad where a FX57 got 118.6 with High Quality settings at 1024x768 using a single 7800gtx. It definately makes one wonder how much of an effect using dual gpus and ATI no less skewed the initial results. I am still using a P4 so like many others am debating upgrade now or wait, so I am really digging hard to find out as much info as I can.

 

[JAG87]

dude you make no sense at all.
first you debate whether the rig was configured for dual core or not, then you debate about dual gpus... so what is the point of this thread, what are you trying to communicate to us here...

my only advice to you, after reading your mayhem is: if you still have a P4, it is time to upgrade, and either way you should wait for AM2/conroe so you dont buy an already obsolete product today.

 

[FallenHero]

yeah classy...i was confused by your post as well. Are you talking about dual gpus, cpus, ati, or nvidia?

 

[Duvie]

So are you saying you dont think the leads will be big in gaming without dual gpus??? I mean most dont run gpus....

This is why I say wait for real reviews...

If you look at the boys over at xtreme I have hardly seen anything other then synthetic benches....superpi, pifast, sciencemark, 3dmock, pcmock, etc.....

I have yet to see many games tested in good cmoparison to an AMD setup out now.

While I have no doubt the Conrope will be better then the AM2, there is too many questions left for sure.....And the lack of helpful results at xtreme (not surprised) doesn't tell me anything we didn't see in AT's original preview....

The only thing Xtreme has helped us with is seeing there is a lot of "no compatability" with current chipset mobos....

 

[classy]

I am no trying to confuse anyone, sorry, my post seems pretty clear. But when I seen the setup Intel used in Bangkok and the scores it got in Q4, it stood out. As I said in my first post, this is the first I have seen of Conroe being used with a single card and the score in Q4 is quite good, but hardly spectacular and makes one question this 20%. Many like me are sitting in wait and maybe being given some smoke and mirrors.

 

[JAG87]

so what are you so confused about... the system with conroe gets 20% more frames. Im still confused as to what you are confused about. Enlighten me will ya!

 

[T2k]

WTF is so confusing? He's absolutely clear: the dual core Conroe with medium settings only 15fps faster than a single core FX-57 with high settings in Q4 or, another occasion, only 18 fps faster than a dual core FX-60 with high settings.

FYI: that 20% is actually 8-11 per cent and only at medium setting, not high. In other words Conroe doesn't look any faster in games.

DIsclaimer: I didn't check his numbers, links.

 

[coldpower27]

The problem is are htese the same settings across all machines as well, dirver revisions as well as what clockspeed what the Conroe benched against the pentium D 950 running at?

 

[classy]

Originally posted by: T2k
WTF is so confusing? He's absolutely clear: the dual core Conroe with medium settings only 15fps faster than a single core FX-57 with high settings in Q4 or, another occasion, only 18 fps faster than a dual core FX-60 with high settings.

FYI: that 20% is actually 8-11 per cent and only at medium setting, not high. In other words Conroe doesn't look any faster in games.

DIsclaimer: I didn't check his numbers, links.

Good disclaimer . The article doesn't say if the game was configured for smp support, but looking at those numbers with a single 7800 gtx it certainly doesn't look to much faster than what we have right now at the high end and the FX 60 that I quoted is using high settings and AA with AF too boot. These are the first numbers I have seen with a single card setup and it certainly raises a couple of questions. Over at xtreme all the numbers they have shown to me are absolutley worthless. I don't understand why they didn't post some real world numbers. Nothing wrong with super pi or 3dmark, but I don't use those programs .

 

[classy]

Originally posted by: coldpower27
The problem is are htese the same settings across all machines as well, dirver revisions as well as what clockspeed what the Conroe benched against the pentium D 950 running at?

Same settings. They don't post what the 950 scored, just that Conroe was faster and the fps it scored in Q4.

 

[JAG87]

Originally posted by: T2k
WTF is so confusing? He's absolutely clear: the dual core Conroe with medium settings only 15fps faster than a single core FX-57 with high settings in Q4 or, another occasion, only 18 fps faster than a dual core FX-60 with high settings.

FYI: that 20% is actually 8-11 per cent and only at medium setting, not high. In other words Conroe doesn't look any faster in games.

DIsclaimer: I didn't check his numbers, links.

ahhhhh alright, I missed that medium and high quality settings stuff. thanks now its crystal clear. next time classy, try not to write your thread into a paragraph, it makes it really hard to read

now for my opinion on this, well... I think that CPUs need to be tested with maxed out graphical power to see if a conroe really makes a difference against an FX60. so a single 7800GTX is not appropriate for this benchmark. the FX57 and FX60 numbers could be lower due to the fact that the 7800GTX sweats with high quality settings. so yes we can estimate that with medium settings the frames are roughly the same as the conroe setup, but then again we probably have a GPU bottleneck.