2 vs 4 cores for Gaming?

[shecklen]

Do cpu's with 4 cores out perform 2 core cpu's with same stock clock speeds when it comes to gaming? Do 4 cores produce more heat? Example: e8400 vs q9650

 

[Helloworld_98]

Most people will recommend a quad core at the moment due to the introduction of them in games, but threads are becoming a bit more important than core numbers.

e.g. there are some 3.46GHz dual cores coming out next year, with turbo mode they'll hit 4.12GHz without overclocking, but they have 4 threads so you'd probably have better performance with one of these in a quad thread optimized app than say a Q9650 OC'ed to 4GHz

 

[shecklen]

I built my gaming rig almost 4 years ago and mobo is ready to go so I'm thinkin of replacing mobo, cpu and memory in the mid cost range instead of building a whole new rig like I have done in the past when things die. I think what ever I choose will be a huge improvement over my amd 4800x2. LOL

 

[neon neophyte]

Most people will recommend a quad core at the moment due to the introduction of them in games, but threads are becoming a bit more important than core numbers.

e.g. there are some 3.46GHz dual cores coming out next year, with turbo mode they'll hit 4.12GHz without overclocking, but they have 4 threads so you'd probably have better performance with one of these in a quad thread optimized app than say a Q9650 OC'ed to 4GHz

what? did you really just say 2 cores with 4 threads will perform better than 4 cores with 4 threads? (under the assumption the program uses all 4 threads.)

im just going to throw this out there.... NO.

 

[Helloworld_98]

^ you're forgetting Nehalem has a higher IPC than Core 2, and the Q9650 is at a lower clock.

It evens the playing field, and look at the benchmarks for i7 vs Q9650, the i7 920 is around the same ballpark, adding clock speed is only going to add performance.

 

[neon neophyte]

even still, 2 cores vs 4 cores? in a 4 thread optimized app? 4 cores wins. (generally speaking, with all cores within the same ballpark for mhz.)

 

[Helloworld_98]

^ but add the extra 2 threads and you'll find them in the same performance level.

 

[neon neophyte]

^ no, you wont. threads do not equal cores.

 

[shecklen]

Excuse my ignorance but what extra 2 threads? I've been out of the game for 4 years and am learning all over again. Does the duel core Cpu have the same amount of threads?

 

[Helloworld_98]

^ GTAIV benchmarks say different, since GTAIV uses 4 threads and the i7 with HT on beats the Q9650 in the GTAIV benchmarks.

 

[neon neophyte]

i7 is 4 cores with 8 threads. of course it performs well with gta4.

the real reason why corei7 outperforms core2quad in apps optimized for 4 threads is because its a real quadcore. core2quads are tremendously difficult to write code that properly utilize all 4 cores.

also, your 8 thread 4 core i7 is not even close to equal of a 8 core 8 thread cpu. which is what this debate was about.

 

[N@n0]

Come on guys...

+1 for neon

*In a 4 threaded app*

clock for clock the 4 core cpu will outperform a multithreaded 2 core CPU, its straight forward.

In a multithreaded 2 core CPU a single core will combine 2 theards of work to be done on one core, while in a 4 core CPU each core handles a thread.

Thus the 4 core CPU (clock for clock) will be able to handel more data/work than a multithreaded 2 core CPU.

*In a 4 threaded app*

Multitasking on the other hand... Nope... 4 core CPU will still be better than a multithreaded 2 core CPU...

These days MHz/GHz isn't everything, and you can't compair a E8400 to a Q9650, they are NOT in the same class. If you compair a E8400 to a Q8400 thats more acurate, but only for 2 vs 4 cores, as these CPUs aren't multithreaded.

As for the heat goes, yes a quad core will generate more heat than a dual core but thats for obvious reasons.

Thats my 2 cents...

 

[leo2kp]

helloworld, try this:

In your BIOS, disable 2 cores and turn HT off and play something or run a CPU bench or 3DMark. Then, drop down to 1 core and turn HT on and play/bench as before.

 

[leo2kp]

N@n0 is right, the number of threads that can be processed simultaneously is misleading. While HT will double the threads that can be processed, it only introduces a slight increase in performance because a particular core is crunching two threads instead of one. Once you hit a point of congestion, performance gains are minimal with HT on. It's better to have multiple dedicated cores than to cram twice the data through one core.

 

[leo2kp]

Furthermore, I have played games where performance has dropped due to HT, and I suspect the reason is that the game is optimized for multiple threads, but instead of using two cores, it will use one core with HT and leave the other three cores idle. Could be wrong :-/

 

[leo2kp]

Excuse my ignorance but what extra 2 threads? I've been out of the game for 4 years and am learning all over again. Does the duel core Cpu have the same amount of threads?

We're talking about HyperThreading, which is a technology where one core will process two threads simultaneously, increasing performance. Not all processors have HT, but the Core i7 does so it can process 8 threads simultaneously vs only 4 threads on a normal quad-core CPU. I don't believe there are any dual-core CPUs out that have HT though, so a dual-core can only process 2 threads at a time.

Our arguement is weather or not having two cores with HT is as effective as having a similarly spec'd quad-core without HT (both having similar architecture and at the same speed). This is easily tested by limiting the cores and disabling/enabling HT on a Core i7.

 

[flyin15sec]

These days MHz/GHz isn't everything, and you can't compair a E8400 to a Q9650, they are NOT in the same class. If you compair a E8400 to a Q8400 thats more acurate, but only for 2 vs 4 cores, as these CPUs aren't multithreaded.

This is incorrect. The Q9x50 is 2 E8x00 glued together. They are the same class.

Q8400 is 2 E5x00(E6300) glued together.

 

[shecklen]

Thanks for the info on HT , threads and cores. It was a good quick lesson to help get me back up to speed.

 

[JAYDEEJOHN]

Well, since duals out duel quads in Ghz, and Nehalem arch is better than C2D, itll be interesting to see how it goes.
Clock for clock, matching the 2 archs, Nehalem wins. Duals will have higher turbo. If they bottleneck at all, it could be, but we really dont have any data , and Im sure itll depend on each game and how it uses the cpu.
Real cores are better, but so is IPC and Ghz

 

[totoaus]

I'd like to contribute from the electronic system design perspective, since that's what I studied in University, back when the original IBM Pc was being intrduced and most of work in multiprocessing was done on the System 370 mainframe. The principles are still valid here, and don't forget that software is part of the system.

My first thought, it that with respect each of you probably has different systems, so the benchmarks are hazy. We large systems people are ALWAYS sceptical of benchmarks, because of limited relevance. I'd love to see someone with the discipline to test a range of configurations on the same machine (where possible). This could be a mac Pro with dual Xeons, max memory & disk in Raid 5 (I'll explain why shortly) and a Windows OS if chips & cores could be selectively disabled while memory, disk, are constant.

Second, any system configuration is bottleneck bound. I have done both business & engineering IT and found that engineers are generally choked by memory bandwidth, while business is choked by I/O bandwidth. I can often see which is causing me troblu on my PC without any performance monitoring tools, and I specified the Mac as I did to hopefully remove memory & I/O bottlenecks for truer CPU performance comparisons.

Now, my expectations are that multiple threaded, mutiple cored and multiple CPU system will show differing performance levels, and they may not be predictable. For example, if the above system were constrained to 1 CPU & 1 thread, a given application will have X performance. It might still only have X performance if the system were fully unconstrained because either the software cannot utilize the extra resources, or both configurations hit a memory or I/O bottleneck, it is ALWAYS application dependent. Another application COULD jump to say 32X performance in the switch 1 CPU, 1 Thread, to 2 CPU, 8 Core, 32 Thread (example, maybe not fact) because it doesn't overwhelm either memory or I/O, yet another run with different parameters could hit a chokepoint and come at much less than 32X.

This is all basic theory, which to same may be well known; but it also appears from your comments that is has been overlooked by many. My point is very simple, what you have been discussing has no absolute answers yet, because it all depends on so many variables.

Now, I found this post in doing research for a friend. It doesn't answer my question though: where can I find a current list of games and other software that can justify the purchase of a quad-core PC over a dual-core PC.

 

[Mousemonkey]

I'd suggest you try starting your own thread on the subject as few people are likely to bother reading your post when they spot the date of the OP.