Best Gaming CPUs For The Money: January 2012 (Archive)

Page 58 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Toggle sidebar Toggle sidebar
Status
Not open for further replies.


Are you sure about that? Because almost every other review of the game proves otherwise:
http://www.gamespot.com/forums/pc-mac-discussion-1000004/crysis-3-is-an-example-of-how-well-amd-cpus-perfor-29356784/
http://i.imgur.com/RhKu6bm.jpg
Those are 3 reviews from three different sources that show similar results against the one review from Tom's you have posted. The i3-4130 is still a hyperthreaded dual core, Crysis 3 is a game that seems to scale far beyond 4 threads and it certainly needs to in order to run as smooth as possible on higher settings.
 
I'm going to keep my 1090T and 965 Phenoms as my top amd builds until Amd brings out something really worth upgrading to. At least with my 1090T and 965 have the Tech that help older games run good too. 3Dnow and AMD64
 

Gamespot forums, lol. And some random Russian site. That must be reliable info.
 
Mate, it's plain obvious that you have very little hands on benching experience with these chips. Claiming that an FX-4100 would beat an Athlon 760K in anything is pure insanity. See my signature? That's a 5.2GHz FX-4100. It loses to the Athlon 760K in everything. I challenge YOU to disable two of your FX-6300 cores and try and out bench my 760K. You will lose, because you are too foolhardy.
 


Are you trolling or are you serious? The gamespot link contains two benchmarks, one done by PCGHX and the other by gamegpu. If you have never heard about either then that is your mistake for living under a rock! And what about the Techspot benchmark as well that I posted earlier and here it is once again just in case you have missed it the first time: http://static.techspot.com/articles-info/642/bench/CPU_03.png. It isn't reliable as well? You seem to dismiss anything that proves against your point as unreliable. That same 'unreliable' random Russian site has just recently benched 'Thief' were the AMD chips including the FX-8350 got beaten by an i3. Would that make it more credible to you? If not why don't you show us your reliable resources?
But I guess you have never known of any other websites that do game benchmarks beside Tom's to start with.
 


It is plain obvious that all you have is claims. Show me some ACTUAL gaming benchmarks which proves your point. Also this article is about best gaming CPU for the money. I don't think pweople who are reading this article will be interested in how much a CPU scores in a synthetic benchmark, that is supposing the 750k destroys the FX-4300 as you claim, since gaming performance shows otherwise. The fact is the A10-5700 is badly beaten in one title by 19 fps difference by even the FX-4100 not the FX-4300!
Edit: I didn't claim that the FX-4100 beat the 760k. I posted several benchmarks that proves that the FX-4100 beats the A10-5700 which is the same as a 750k with disabled iGPU.
 

If it takes a 5 GHz AMD CPU to beat a 3.4 GHz Intel CPU, that's not a whole lot to brag about.
 

Instead of boosting about your great overclocking and synthetic benchmarks achievement, referring to your profiles on other sites and your accolades, ridiculing other people's opinions and knowledge in such a condescending manner in your two last replies, you could have provided any gaming benchmark results to support your point, which you have failed to do, I don't care if the 750k/760k score higher than the FX-4300 in synthetic benchmarks, you claimed that even a FX-63xx or a FX-83xx is not a better gaming chip than the A series APUs and their counterpart Athlon chips, I provided benchmark results that proves otherwise, some of them you have arrogantly dismissed as sloppy, so I provided four other benchmarks from a different source, in two of them the FX-4100 which is a bulldozer chip with weaker cores scored 11 and 19 fps higher than an A10-5700 in Metro Last Light and Splinter Cell Black List respectively, The FX-4170 beats the A10-5800k by 7 fps in Crysis 3, The FX-4170 wins by smaller margin in Far Cry 3 still beating the A10-5800k and the FX-6300 wins by 7 fps difference. I don't know really whether the L3 cache impacts performance that much or not, but these results prove if anything that claiming that the FX-6300 and other FX chips are not any better as gaming chips than the 750k/760k and that the FX-6300 shouldn't be recommended over the Athlon chip is simply not true.

Oh and as a side note, being such a knowledgeable expert you failed to explain why you deliberately handicapped the FX-4300 rig with slower memory, 150 MHz lower clocked GPU and lower clocked graphics memory in the 3dmark physics benchmark you have posted, the FX-4300 was also clocked a little slower if I remember correctly, and you did not answer my question regarding the validity of the Physics score which seems to have a certain degree of GPU dependency especially with AMD GPUs. When taking the previous factors in considerations it is not a surprise that the Athlon chip scored 'slightly' higher than the FX-4300. If you really wanted a good example of the trinity chip winning by a [strike]relatively good lead over the FX-4300 at the same clock speed you could have posted Cinebench results instead[/strike] (scratch that, actually the A10-5800k merely achieves 0.01 point higher score over the FX-4300 at the same clock speed in Cinebench), which is still irrelevant to the subject of this article, since the only thing that matters here is how these chips bench in games.
 

I'd expect the 760K to beat the FX-4100, but not the FX4300 (which is what I assume you mean by disabling two cores on a 6300; OC it to 3.8).

But your avatar here and elsewhere is probably going to get you into trouble, with that attitude.
 

Not to take sides in this "pissing contest", but can you explain exactly how being ranked in the "top 10% in the world" for overclocking a CPU even remotely relates to understanding system design, performance, and overall value of a PC gaming system? I think you'll find that trying to brag about those credentials on these forums won't win you many advocates.
 
For my part, Damric, it looks like you have done quite a bit of testing, of a great many systems. The issue here though seems to be the very specific reference to gaming, and a general disdain of synthetics. I'm not going to challenge or support the value of synthetics, but as many tests as you have run, would it be too much to ask for some gaming-only benchmarks of the chips under discussion? While you may have entirely valid reasons for being certain of the results, would it be too much to ask to humor the doubters with a few specific tests?
At the high end, differences of 10-20 FPS may hardly be noticed, but at the low end, where these budget chips live, a 5 FPS difference may be quite visible.
 

Intel used to do the same thing in the NetBurst days until they hit the 3.8 GHz wall and they couldn't even get it stable at that frequency. Chances are slim that AMD is going to drop that 'module' architecture though the same way Intel dropped NetBurst and went back to a tweaked version of their previous microarchitecture.
 

Netburst had three distinctive features:
1- uOP cache ("trace cache")
2- long pipeline
3- hyperthreading

The uOP cache got reused in Core2, Core-iX and derived products. The long pipeline got shortened since it failed to yield the expected timing budget for higher clocks and was wasting too much power for the benefits it failed to produce. Hyperthreading is still with us in various SKUs ranging from Atom to Xeon.

The only thing Intel really ditched when they "went back to what worked" is the long pipeline.
 
Only Pentium 4. Pentium D didn't have hyperthreading. I hope that we start seeing some HSA implementations on the software side soon. Fusion won't mean anything without them.
 


Here, I tried to find a "gaming" benchmark that was free to download for you. Now I didn't dial in very high overclocks, just in case you can't clock that high yourself. So let's see those 4 core FX chips with the 8mb L3 cache come in and tear it up shall we?

LL


If you know of any other free "gaming" benchmarks that you would like to compare, let me know.

 
I'm not the one who needs convincing. If my system(s) run my games on acceptable settings and good FPS, I'm not worried about it. I've thought about building a "how low can you go" test PC though, just for grins; likely on FM2+ because it looks like AM1 is just too weak.
 

The Pentium-D is just a pair of stand-alone Prescott P4 cores connected to the FSB over the CPU substrate. The early Pentium-D were two physically separate Prescott cores while the later Pentium-Ds were two consecutive dies cut from the same wafer as one. Since Prescott had much worse TDP than Northwood, Intel could not afford the thermal headroom to leave HT on in the Pentium-D.

The Willamette P4 did not ship with HT enabled either but its Xeon variant and some engineering samples did.

The Core/Core2 did not have HT at all but that was probably just Intel not wanting to mess around with too many things at once and heavily focusing on power-efficiency during that product generation.
 


Interesting... Apologies if it is there as the image is too small to even read your processor model, but what kind of FPS rates does that score translate to?
 

Which is why I switched from a PIII Katmai on my Win98 machine to an Athlon XP T-Bred when I built my WinXP box ( sadly I couldn't afford the Coppermine in the Win98 days. ) AMD got its collective stuff together with the T-Bird and I liked that the Palomino chips continued that. The parts were less than the Intel offerings, I didn't have to deal with the heat from the P4, and I didn't lose any performance at all. My nForce board also unlocked the CPU multiplier so when it started to lag behind around WinXP SP3, OCing gave it some extra life. When I got a laptop in 2004, I got a Pentium M because that had the best performance/battery life offering at the time.

I build systems based on efficiency at a set performance threshold, not name brands.
 


It equates to nothing, as I was like 3x the FPS of my refresh rate of my monitor. It's just for comparison to people that think the lack of L3 cache on the fm2 Athlons means anything. When I posted futuremark benchmarks, they complain it's synthetic. Well just about every CPU Benchmark is synthetic.

 

Fwiw, I am inclined to agree with you, provided enough synthetics provide the same results. A single synthetic may not tax a CPU anywhere near the way a game (or other specific application) might, but many of them showing similar, consistent results seems difficult to refute.
 
His Athlon and overclocked HD 7850 are beating an i5 3570k @ 4.5 GHz and a GTX 660 Ti:
10


900x900px-LL-1562f083_wwww.jpeg

here is the page were the guy with the i5 posted his benchmark results:
http://www.overclock.net/t/1363297/square-enix-final-fantasy-xiv-a-realm-reborn-official-benchmark-exploration/310#post_20814842
You will find similar results, the benchmark seems GPU bound more than anything else. Take a look at the SLI and Crossfire setups and you will see them exceeding the 20 k mark. I find it strange that you think that synthetic benchmark results are more valid than in game bench mark results. Shouldn't the latter be more difficult to refute especially since we are talking about gaming chips here? Anyway no point of continuing this argument anymore.
Edit: This is in case the above images don't show up:
http://www.overclock.net/t/1363194/final-fantasy-14-a-realm-reborn-benchmark-2-0/10
http://cdn.overclock.net/1/15/900x900px-LL-1562f083_wwww.jpeg
 


I think you are missing the point here: The easiest general representation of gaming performance on a system is FPS. A system pulling 60 FPS is going to feel a heck of a lot better in a given game than a system pulling 30 FPS in the same game with the same settings. If you are claiming that a 750K is a better gaming processor than a specific FX model processor then you are saying "If you take all of the parts in system X with a 750K processor, run some gaming benchmarks, and then compare those benchmark results to an identical system with an FX-XXXX processor as the only difference then the system with the 750K will have better results as demonstrated by the FPS results of the benchmarks."

I used to bench my system with Aquamark, which gave an overall score like your FF benchmark but also included the FPS results, but that was a decade ago and I don't think that benchmark will even run on today's systems.

In other words, what you need to understand is that everyone else is pointing out that synthetic results don't always transfer to real-world gaming performance. Therefore, backing up your claims that your Athlon processor beats FX processors in gaming (which apparently flies in the face of what everyone else has experienced) means posting links to benchmarks of identical systems with only a change in the CPU that shows the Athlon system pulling better frame rates than the FX system.
 
Status
Not open for further replies.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom