• 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.

Thinking of switching from Intel to AMD

You have to go back a few years before you get 2 threaded games being common. And most of those are nowhere near enough demanding to have a problem on an 8350. Once you get to 4 cores things tend to level up, and when you get to 6-8 cores, the 8350 tends to pull ahead.
 
Actually you are missing the obvious. Selling the 2500K and mb\cooler\mem will pretty much pay for the Amd system.
AMD would still need the cooler and mem, so keeping those, selling the i5 2500K and going for FX8320 is +£20, and then the Z68 board struggle to fetch decent money (because of Z87 and Haswell)....best case scenerio fetching £50 for the board, and getting a AM3+ board at say...£100-£120, that would be £70-£90 extra in total.

Got to disagree there :p
Yea...the only well-threaded games are the only odds titles from EA...and few from Ubisoft I think (and I think even AC 3 was a 1-2 threads bugger? Only really recall Far Cry 3 being well threaded beyond 4 threads, but the game itself was buggy as hell).
 
Last edited:
But you need more threads used than an i5 has cores to be ahead.

Yes and there are games out there (i.e. list previously posted) which use more than 4 threads. I suspect this is the type of info the OP would be more interested in.

If he buys a CPU which runs multi-threaded games well now and in the future, he doesn't have to worry about bottlenecks in future especially with more powerful GPUs/multiple GPU setups. Mantle hopefully will only enhance this meaning you can stick to one CPU while upgrading GPUs for years. The CPU upgrade point I suspect will come around faster if he has a 2500k than if he has an FX CPU. That's all we are saying.
 
Proof? People also expected PS3 and Xbox360 with there 6 and 7 CPU threads to rapidly increase the speed at which games become threaded... We all know how that turned out.

This year so far has shown otherwise.

Reviews have show it to run on up to 6 threads and I've seen it myself.

Unless it runs very differently in game to its own benchmark, I don't believe that. Either that or somehow by magic, they made it only use 2(ish) cores on an AMD FX CPU. Links to these reviews?
 
Unless it runs very differently in game to its own benchmark, I don't believe that. Either that or somehow by magic, they made it only use 2(ish) cores on an AMD FX CPU. Links to these reviews?

http://forum.beyond3d.com/showpost.php?p=919807

It runs anything from 5-7 threads on console and if ypu search google for performance numbers many sites report a good increase in frame rate when running from a dual core to a quad core..
 
http://forum.beyond3d.com/showpost.php?p=919807

It runs anything from 5-7 threads on console and if ypu search google for performance numbers many sites report a good increase in frame rate when running from a dual core to a quad core..

I can only conclude that it has been coded poorly/deliberately to utilise less cores on AMD FX CPUs.

Martini and Teppic did you see how many threads were used in your benchmark?
 
Here's my resi6 bench with my 24/7 settings (i.e. CPU at 4.8 GHz, 7990 at 1100MHz core). Could only see 2 cores being utilised to any decent extent:


I scored just under 11000 points with a single 7950 at 1080p maxed out, and you only manage an extra ~3000 points with double the GPU power?

You are CPU limited, at the latter stage of the first benchmark sequence my 7950 dropped down to 82% use, apart from that it was pegged at 99% use throughout the whole benchmark.
 
I scored just under 11000 points with a single 7950 at 1080p maxed out, and you only manage an extra ~3000 points with double the GPU power?

You are CPU limited, at the latter stage of the first benchmark sequence my 7950 dropped down to 82% use, apart from that it was pegged at 99% use throughout the whole benchmark.

Not disputing this benchmark makes the FX chips CPU-limited. We have already gathered that. I said myself it only uses 2 cores, so with lower IPC compared to Intel CPUs of course this will happen.

Run with both your cards. You won't get double the score though. It doesn't work like that. You may get around 16k. My 3930k system on 24/7 settings (at the settings in the sig) gets 19200 score.
 
Seriously, GTFO with those pathetic excuses...

Honestly..... :rolleyes:

Excuses? Are you high? Why would I need to make excuses for anything. I saw it use 2 cores during the benchmark. On the other hand, you pretend to verify that the resi6 game engine is well threaded and you post some link from 2007 to do this and tell me to search google for the PC reviews that show this.
 
Martini and Teppic did you see how many threads were used in your benchmark?

One hit very hard, the rest not very much, a couple barely used at all.

I didn't get that much higher than you, considering I've got nvidia SLI in an nvidia benchmark and a 4770K at 4.5GHz.
 
Excuses? Are you high? Why would I need to make excuses for anything. I saw it use 2 cores during the benchmark. On the other hand, you pretend to verify that the resi6 game engine is well threaded and you post some link from 2007 to do this and tell me to search google for the PC reviews that show this.

"Resident Evil 6 is powered by the MT Framework engine, and we were really impressed with its multi-core support as there was a 20-30fps difference between a dual-core and a quad-core CPU. The only game that benefited that much from two additional CPU cores was Battlefield 3 and MOH: Warfighter, two games that were powered by DICE’s Frostbite 2 Engine. It’s pretty obvious that Resident Evil 6 scales well on all four cores and there was a 70-80% CPU usage, so kudos to Capcom for its engine’s optimization. The good news is that Resident Evil ran with constant 60fps on both our dual-core and our quad-core systems in Single-GPU mode (at 1080p with max details and we should note that our GPU was not stressed most of the times), meaning that PC gamers will be able to enjoy it even if they don’t own a top of the line high-end PC system."

http://www.dsogaming.com/pc-performance-analyses/resident-evil-6-pc-performance-analysis/

Learn to use Google....
 
"Resident Evil 6 is powered by the MT Framework engine, and we were really impressed with its multi-core support as there was a 20-30fps difference between a dual-core and a quad-core CPU. The only game that benefited that much from two additional CPU cores was Battlefield 3 and MOH: Warfighter, two games that were powered by DICE’s Frostbite 2 Engine. It’s pretty obvious that Resident Evil 6 scales well on all four cores and there was a 70-80% CPU usage, so kudos to Capcom for its engine’s optimization. The good news is that Resident Evil ran with constant 60fps on both our dual-core and our quad-core systems in Single-GPU mode (at 1080p with max details and we should note that our GPU was not stressed most of the times), meaning that PC gamers will be able to enjoy it even if they don’t own a top of the line high-end PC system."

http://www.dsogaming.com/pc-performance-analyses/resident-evil-6-pc-performance-analysis/

Learn to use Google....

I should learn to use google to "prove" a claim you have made? Don't think it works like that actually.

That "review" you posted shows absolutely no evidence of decent multi-core/multithreaded performance. Just a wall of text and a couple of irrelevant screenshots. No comparative data although there is a screenshot showing 4 cores being utilised but no idea if that is within game or on the bench or if it even is in Resi6

In the real world, a few of us have now run this benchmark and we are not seeing 70-80% usage in 4 cores. Go figure. Unless they forgot to optimise the bench engine and the actual game uses 4 cores. Who knows.
 
Last edited:
Back
Top Bottom