• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

3 Core Phenom is real

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.
The first link in this reply wasn't part of what you originally linked to me when i first replied in this thread based on clock for clock scaling while in a GPU bound scenario.

I know, I just showed the post so you can see I was aware that the bench wasn't cpu bound, but because UT doesn't even load my cpu to the fullest ingame I felt that image was enough, the small difference between the 2 cpus might be when occasionally a K8 is maxed out.

Yes the flyby is stressing the processor heavily however what it's showing for results does not merit the absence of physics/bots since in the realworld, there will always be physics/bots or players if you are playing mutli.

Although it does not represent a real gameplay it puts more stress on the cpu than usual gaming res so it's closer to reality than a high res flyby.

Depending on personal preference, this is debatable. I personally do not like my games to dip anywhere below 50fps. I fully agree that the difference from 80 and 160 is negligible for AVG framerates however the real determining factor is the personal preference of how far you mind the framerates to dip. Anandtech simply does not display this difference unlike PcPerspective and HardOCP which if I am really interested in performance then I will goto either of those sites. I will also ask people in forums.

Sure thing, that's why I wrote 80 as long it doesn't fall under 40 I don't care if it's an FPS for RTS it can go even lower.

Keep in mind that this demo is not really testing the full capabilities of the full game. You just aren't going to know for a fact what performance you will get until it is released.

We will see, I have roughly 20% for the changes.

You are using UT3 as your main example fordual vs quad in PC gaming which is bad mainly for the fact that not all games are the same. Like I mentioned in my other posts, there are games out now that do show a clear difference in playability when comparing duals to quads like Supreme Commander and World in Conflict. The larger the map with the more units will just bog any dual to shreds. It's silly to just recommend a dual core only because your gaming preference isn't derived around such games out now that don't fully utilize quads. Not everyone likes FPS gaming.

No, there wasn't anything about quads in your post to what I replied, I used it in C2D VS slower duals E2XXX or K8 comparison, you compared X2 and E6XXX processors, that's why there wasn't any quad on the results I posted.

See for yourself.

OBLIVIONLORD said:
If they even have the nerve to price these chips up High like they are doing to the current x2 6000 and 6400 then I'll lose even more respect for them. The 90w version of the 6000 shouldn't be anymore than E6550 since they are both neck and neck for performance even the E6650 is still more energy efficient. The 125w version should be even less but, it's a mere $10 less at newegg then the E6550 for 35w more power compared to the 90w version.

It still stands, I do not need a C2D...

I used UT because that was the most demanding I encountered so far, could say let's bench Crysis but then you will only see 60-80% cpuload which makes it even more hard to explain how would I benefit from a C2D.
I also thought RTS with the million units will be the best to put my cpu to the test but the load Supreme Commander and WiC put on the cpu in the campaign or 4 player multi wasn't even significant enough to mention compared to UT3.

By the way crysis devs said their engine needs cpu more than gpu.

hack: What is the main limiter for Crysis in terms of GPU, CPU, or RAM? If users are near the low end of the requirements, which should they upgrade first?

Cevat Yerli: We would say first CPU, then GPU, then memory. But it must be in balance. If you are balanced, we are more CPU bound then GPU, but at the same time at higher CPU configurations we scale very well for GPUs.

So there it is, I ran that game which is more CPU than GPU bound with 60-80% load, do I want a faster C2D ? No.
Do I need or want a faster clocked C2D ? No.
A quad ? Not yet, by the time I will really need one there gonna be some really good deals and steppings probably even a Nehalem.
 
Although it does not represent a real gameplay it puts more stress on the cpu than usual gaming res so it's closer to reality than a high res flyby

It's not closer to reality at all. It's purely a flyby test so regardless if it stresses the cpu or not at a high or low res, it does not represent any "Real" results. Just because the cpu is being loaded does not mean that your going to get more FPS than a faster processor that's equally loaded. Plain and simple.

I also thought RTS with the million units will be the best to put my cpu to the test but the load Supreme Commander and WiC put on the cpu in the campaign or 4 player multi wasn't even significant enough to mention compared to UT3.

These 2 games load much more than UT3. UT3 mainly emphasizes on GPU not CPU. These games both have a simulation thread which the more units the more stress. Considering that each individual unit is calculating their on commands which a FPS will only do depending on the amount of bots you have in game then you are quite limited to really stressing UT3 demo mainly because you can't use bots for any sort of real world testing.

If you really had 4000 units in total in SC on the largest map then I highly doubt you are truthful when saying that the load "wasn't even significant" compared to UT3.

No, there wasn't anything about quads in your post to what I replied, I used it in C2D VS slower duals E2XXX or K8 comparison, you compared X2 and E6XXX processors, that's why there wasn't any quad on the results I posted.

I jumped the gun when I replied because I made so many edits thinking that I mentioned quads but, regardless. You did make mention to quads in your last post.....

By the time quads will be a requirement to run games properly the E6XXX and the E2XXX line will go down just alike so I do not expect those midrange procs to last significantly longer nor to justify the price.

All I simply did was reply that "some" modern games properly utilize each core therefore making quads for current gaming a greater value than dual. It only gets better in the future when more companies disperse threading processes around more than just 2 cores.
 
Last edited:
Back