Choosing a Gaming CPU: Single + Multi-GPU at 1440p, April 2013
by Ian Cutress on May 8, 2013 10:00 AM ESTCPU Benchmarks
Point Calculations - 3D Movement Algorithm Test
The algorithms in 3DPM employ both uniform random number generation or normal distribution random number generation, and vary in amounts of trigonometric operations, conditional statements, generation and rejection, fused operations, etc. The benchmark runs through six algorithms for a specified number of particles and steps, and calculates the speed of each algorithm, then sums them all for a final score. This is an example of a real world situation that a computational scientist may find themselves in, rather than a pure synthetic benchmark. The benchmark is also parallel between particles simulated, and we test the single threaded performance as well as the multi-threaded performance.
As mentioned in previous reviews, this benchmark is written how most people would tackle the situation – using floating point numbers. This is also where Intel excels, compared to AMD’s decision to move more towards INT ops (such as hashing), which is typically linked to optimized code or normal OS behavior.
Compression - WinRAR x64 3.93 + WinRAR 4.2
With 64-bit WinRAR, we compress the set of files used in our motherboard USB speed tests. WinRAR x64 3.93 attempts to use multithreading when possible and provides a good test for when a system has variable threaded load. WinRAR 4.2 does this a lot better! If a system has multiple speeds to invoke at different loading, the switching between those speeds will determine how well the system will do.
Due to the late inclusion of 4.2, our results list for it is a little smaller than I would have hoped. But it is interesting to note that with the Core Parking updates, an FX-8350 overtakes an i5-2500K with MCT.
Image Manipulation - FastStone Image Viewer 4.2
FastStone Image Viewer is a free piece of software I have been using for quite a few years now. It allows quick viewing of flat images, as well as resizing, changing color depth, adding simple text or simple filters. It also has a bulk image conversion tool, which we use here. The software currently operates only in single-thread mode, which should change in later versions of the software. For this test, we convert a series of 170 files, of various resolutions, dimensions and types (of a total size of 163MB), all to the .gif format of 640x480 dimensions.
In terms of pure single thread speed, it is worth noting the X6-1100T is leading the AMD pack.
Video Conversion - Xilisoft Video Converter 7
With XVC, users can convert any type of normal video to any compatible format for smartphones, tablets and other devices. By default, it uses all available threads on the system, and in the presence of appropriate graphics cards, can utilize CUDA for NVIDIA GPUs as well as AMD WinAPP for AMD GPUs. For this test, we use a set of 33 HD videos, each lasting 30 seconds, and convert them from 1080p to an iPod H.264 video format using just the CPU. The time taken to convert these videos gives us our result.
XVC is a little odd in how it arranges its multicore processing. For our set of 33 videos, it will arrange them in batches of threads – so if we take the 8 thread FX-8350, it will arrange the videos into 4 batches of 8, and then a fifth batch of one. That final batch will only have one thread assigned to it (!), and will not get a full 8 threads worth of power. This is also why the 2x X5690 finishes in 6 seconds but the normal X5690 takes longer – you would expect a halving of time moving to two CPUs but XVC arranges the batches such that there is always one at the end that only gets a single thread.
Rendering – PovRay 3.7
The Persistence of Vision RayTracer, or PovRay, is a freeware package for as the name suggests, ray tracing. It is a pure renderer, rather than modeling software, but the latest beta version contains a handy benchmark for stressing all processing threads on a platform. We have been using this test in motherboard reviews to test memory stability at various CPU speeds to good effect – if it passes the test, the IMC in the CPU is stable for a given CPU speed. As a CPU test, it runs for approximately 2-3 minutes on high end platforms.
The SMP engine in PovRay is not perfect, though scaling up in CPUs gives almost a 2x effect. The results from this test are great – here we see an FX-8350 CPU below an i7-3770K (with MCT), until the Core Parking updates are applied, meaning the FX-8350 performs better!
Video Conversion - x264 HD Benchmark
The x264 HD Benchmark uses a common HD encoding tool to process an HD MPEG2 source at 1280x720 at 3963 Kbps. This test represents a standardized result which can be compared across other reviews, and is dependent on both CPU power and memory speed. The benchmark performs a 2-pass encode, and the results shown are the average of each pass performed four times.
Grid Solvers - Explicit Finite Difference
For any grid of regular nodes, the simplest way to calculate the next time step is to use the values of those around it. This makes for easy mathematics and parallel simulation, as each node calculated is only dependent on the previous time step, not the nodes around it on the current calculated time step. By choosing a regular grid, we reduce the levels of memory access required for irregular grids. We test both 2D and 3D explicit finite difference simulations with 2n nodes in each dimension, using OpenMP as the threading operator in single precision. The grid is isotropic and the boundary conditions are sinks. Values are floating point, with memory cache sizes and speeds playing a part in the overall score.
Grid solvers do love a fast processor and plenty of cache in order to store data. When moving up to 3D, it is harder to keep that data within the CPU and spending extra time coding in batches can help throughput. Our simulation takes a very naïve approach in code, using simple operations.
Grid Solvers - Implicit Finite Difference + Alternating Direction Implicit Method
The implicit method takes a different approach to the explicit method – instead of considering one unknown in the new time step to be calculated from known elements in the previous time step, we consider that an old point can influence several new points by way of simultaneous equations. This adds to the complexity of the simulation – the grid of nodes is solved as a series of rows and columns rather than points, reducing the parallel nature of the simulation by a dimension and drastically increasing the memory requirements of each thread. The upside, as noted above, is the less stringent stability rules related to time steps and grid spacing. For this we simulate a 2D grid of 2n nodes in each dimension, using OpenMP in single precision. Again our grid is isotropic with the boundaries acting as sinks. Values are floating point, with memory cache sizes and speeds playing a part in the overall score.
2D Implicit is harsher than an Explicit calculation – each thread needs more a lot memory, which only ever grows as the size of the simulation increases.
Point Calculations - n-Body Simulation
When a series of heavy mass elements are in space, they interact with each other through the force of gravity. Thus when a star cluster forms, the interaction of every large mass with every other large mass defines the speed at which these elements approach each other. When dealing with millions and billions of stars on such a large scale, the movement of each of these stars can be simulated through the physical theorems that describe the interactions. The benchmark detects whether the processor is SSE2 or SSE4 capable, and implements the relative code. We run a simulation of 10240 particles of equal mass - the output for this code is in terms of GFLOPs, and the result recorded was the peak GFLOPs value.
As we only look at base/SSE2/SSE4 depending on the processor (auto-detection), we don’t see full AVX numbers in terms of FLOPs.
Facebook
Twitter
RSS
242 Comments
View All Comments
CookieKrusher - Saturday, May 11, 2013 - link
Good to know that my 2500K is still taking care of business. Now if I could just upgrade this GTX 460 this year, I'll be golden. :-)Tchamber - Saturday, May 11, 2013 - link
I wonder why games are vastly more parallel on the GPU side of things than the CPU side. If a game can utilize 2048 SPs, why doesn't adding 2 or 4 more CPU cores help much?ShieTar - Tuesday, May 14, 2013 - link
Because all parts of the code which can be run in parallel are already running on the GPU, and the CPU is stuck with the code that needs to be serial.MelodyRamos47 - Sunday, May 12, 2013 - link
Sienna. true that Amber`s artlclee is impossible... last saturday I bought Ariel Atom after having made $9498 this-last/5 weeks and-also, $10,000 last-munth. with-out any question its the coolest work I have ever had. I started this seven months/ago and pretty much straight away started bringin home minimum $71 p/h. I follow the details on this straightforward website, Bow6.comTAKE A LOOKOwnedKThxBye - Sunday, May 12, 2013 - link
Love this information it was an eye opener. Great Job Ian!To choose a gaming CPU is a question I am asked to answer nearly on a daily basis from clients or friends in my line of work. While your concluding recommendation are spot on given the information you provided, I wouldn't often find myself giving out the same advice. The reason behind this is the future upgrade path of the PC. My apologies if this has already been pointed out in the comments as I haven’t read every one yet.
Most people seeking a PC upgrade have just started playing a new title and have hit a wall. They are unable to play this new game at the resolution and detail they feel to be the minimum they can put up with. This wall is mostly a CPU or GPU limitation but sometimes it’s both. Of these upgrades the new graphics card is significantly less expensive than a full system upgrade, can be installed easily by most people, and doesn't leave you without a PC or any down time. On the other hand a full system upgrade is expensive, not everyone can put it all together, and often requires an OS reinstallation with data backup.
Let’s say an average gamer (not necessarily you and me) purchases a nice new gaming rig today for whatever reason. It’s likely that within two years or so they are going to hit a wall again. At this point most people have hit the GPU limitation and are able to upgrade the graphics card and away they go again for another one to two years. After hitting this wall for the second time it’s most likely time for a full system upgrade. This process could be only two years for some of us but for others it’s going to be four to five.
What I’m trying to point out is that we can recommend a CPU that is the cheapest while still not limiting our current GPU and get the best possible FPS per dollar right now. But if we do this it’s far more likely we are going to run into a CPU bottleneck early in the upgrade path and instead of forking out a few hundred for a new graphics card after a year or two, we might end up having to replace the both the CPU and motherboard as well.
For this reason I could not recommend an AMD A8-5600K or an equivalent Intel CPU to be purchased with a HD7970 or GTX580 unless you plan to never upgrade your graphics card. Spend the extra $100 to $150 on a better CPU and potentially make the PC last another two years. Maybe the inclusion of some popular titles like Battlefield 3 or PlanetSide 2 would have significantly changed your concluding recommendations. The information provided gives us a good indication of where the CPU bottleneck comes into play but I think the upgrade path of the PC along with what games are being played need to be given a lot more weight for an accurate recommendation to be made. Having said that I could be totally wrong and have recommended the wrong gaming builds for years.
TheJian - Monday, May 13, 2013 - link
I can see a lot of work but only for a future that won't exist for a good long while. You tested at a res that is too high and not showing reality today based on this dumb idea that we'll all buy $400+ monitors. This is the same crap Ryan tries to push (see the 660ti comments section, he pushed it then when they were $600 and ebay via your visa to korea...ROFLMAO - read as I destroyed his responses to me, click ALL comments so you can just CTRL-F both of us). So raise your hand if you're planning on buying a $400+ monitor, to go with an expensive $300 card only to end up at say 28fps in a game like sleeping dogs (avg...so game is unplayable as minimums would be oh I don't know 15fps?). I don't see any hands raised. So we'll be lucky if MAXWELL in Q1 (or whatever Volcanic does for AMD Q4 this year) will be playable at 1440p. Translation, we'll all buy 1920x1200 or 1080p for a while to come unless we own more than one card. Raise your hand if you have multi-gpu's. According to steampowered.com hardware survey that number (last I checked) was under 2%. You're wasting your time. Start writing for the 98% instead of the 2%. I just wasted MY time reading this crap.REALITY: We are all gaming at 1920x1200 or 1080p (or worse, below this). This should be the focus. This would show LARGE separations in cpus and Intel kicking the crap out of AMD and that you wouldn't want to touch that A8-5600 with a 10ft pole. Why? The 7970 would not be the limiter, or at least not every time like here. What % of the people have 3-4 gpus? Give me a break this is what you see as the future? $1200 in gpus and a $400+ monitor? You're pandering to a crowd that almost doesn't exist at all. For what? To make an AMD cpu seem buy-able?
The data in this article will be useful in 3yrs+ when you can hit 1440p at above 30fps MINIMUM on most cards. Today however, we needed to see what cpu matters at a resolution that doesn't make a 7970 look like a piece of outdated trash. You're pretty special today if you have 7970 or up in the gpu.
More AMD CYA if you ask me. Just like we're still waiting months for Ryan to do an FCAT testing article...LOL. We'll be waiting for that until xmas I'd guess unless AMD doesn't get the prototype driver done by then, which means we'll never see FCAT testing here...ROFL.
Ryan has ignored TWO articles now on fcat. It didn't make the 7990 review, and part2 of fcat article never even came. Just keep delaying, your sites credibility is going down the drain while everyone else tells it like it is. AMD & their drivers currently SUCK (cpu & gpu). Their cpu's suck; hence running at a res that shows all your games can't run without multi-gpu and hit 30fps+ MINIMUM - meaning at this res they ALL require more than one gpu making cpu choice a non issue of course. Their gpu's are great but drivers suck so they give away games by the truckload to try to sell a gpu vs. exceptional NV drivers. Lets face it, the best hardware in the world sucks if drivers can't live up to the hardware. Unfortunately AMD blew all their R&D on consoles that are about to die on the vine, instead of GREAT drivers to go with a GREAT gpu.
What do you end up with when you spend your wad on consoles instead of drivers? FCAT showing you suck, runts, stutter, enduro that lacks on notebooks (see notebookcheck 7970m article recently, it was even mentioned here oddly...LOL) and CF that is abysmal and at times showing NEGATIVE scaling for more than one gpu vs....again, NV drivers that have none of these issues. Optimus works (hence nv beats this drum repeatedly and justifiably) and so does their SLI. While AMD sucked for a year (see hardocp driver review for AMD & NV recently) NV got to sit on their butts (driver butts) waiting for AMD to finally get done with consoles and make a "Never Settle" driver that actually performed the way the cards should have OUT OF THE BOX! Thank god for never settle drivers in Nov or Nvidia wouldn't have released monthly driver enhancements from Dec-May...ROFL. People would be stuck with the same perf today as out of the box from NV (as hardocp showed they didn't improve 1fps all year until AMD caught them...well duh, no point giving out free perf when blowing your enemy away all year).
Mark my words...AMD will be writing off R&D for consoles soon. Even activision's Kotick just said last week that consoles (for all the reasons I've said repeatedly here and at tomshardware etc) have a very tough road ahead vs. all the new ones coming out. Sales of Wiiu off 50% after xmas pop. Just one Q after debut nobody cares already! He basically said they'll be watching for the same on the next two (ps4/xbox720). When that happens no games will be made going forward for this crap as we all move to tablet/phone/ cheaper console or PC (for ultimate gaming).
Video killed the radio star. Cheap android gaming killed the console star....LOL.
Ouya, Steambox, Shield (pc to tv here!), wikipad, razer edge, gamepop etc...All stuff that will help kill the consoles and stuff they have never faced before. It was always the big 3, but his time big 3 with little 6-10+a billion phones & tablets chasing them and our gaming time...ROFL. The writing has been on the wall for a LONG while. As usual AMD management screws up. Wisely NV passed on a dying market and only spent 10mil on both Shield and Grid respectively...ROFL. Dirk Meyer wouldn't be doing this crap. They were idiots letting him go thinking he didn't get it. He had a mobile strategy, it just wasn't one that made their CORE products suck while creating it. Management has PIPE dreams. Dirk had REALITY dreams.
http://www.tomsguide.com/us/Next-Generation-Bobby-...
Kotick saying consoles are dead, well he almost says it...Wait and see is basically the same thing...LOL.
"If I were gaming today on a single GPU, the A8-5600K (or non-K equivalent) would strike me as a price competitive choice for frame rates, as long as you are not a big Civilization V player and don’t mind the single threaded performance. The A8-5600K scores within a percentage point or two across the board in single GPU frame rates with both a HD7970 and a GTX580, as well as feels the same in the OS as an equivalent Intel CPU."
AMD CYA. Total lie. Drop this crap down to 1080p and watch the Intel chips separate the men from the boys and in MORE than just CIV5. ALL games would show separation I'd guess. You must of found this out, which immediately made you up the res huh? AMD threaten the free money or something if you showed them weak or Ryan managed to run FCAT testing?...LOL.
"We know the testing done here today looks at a niche scenario - 1440p at Max Settings using very powerful GPUs. The trend in gaming, as I see it, will be towards the higher resolution panels, and with Korean 27" monitors coming into the market, if you're ok with that sort of monitor it is a direction to take to improve your gaming experience."
Seriously? "If you're ok with EBAYing your $400 "KOREAN" monitor this is a great way to improve your gaming at under 30fps minimum in all games...ROFL. Reworded for clarity Ian :)
NICHE situation is correct in that first sentence...LOL. Again, start paying attention to your audience which is 98% not the NICHE 2% or less. I'm debating buying a 2nd 1920x1200 (already have 2 monitors, one dell 24 and a 22in at 1680x1050) instead of your NICHE just because of what you showed here. 1440p is going to be difficult to run ABOVE 30fps MIN for ages. I'd spend most of my gaming time on the smaller dell 24 at 1920x1200 I think. So I'm debating buying the same thing again in 27in. I want a bigger screen, but not if I can't run 30fps for another 2-3 vid card revs (maxwell rev2?). This is just like I described above with AMD's gpu. Great hardware, but worthless without drivers that work right too. A korean monitor may look great, but what is it worth if you require $600+ in vid cards to have a prayer of 30fps? I'd rather buy a titan, not upgrade the monitor and hit well above 30fps on my dell 24 at 1920x1200 all day! THAT is a great gaming experience I can live with. I can't live with a BEAUTIFUL SLIDE SHOW on a korean monitor off ebay...LOL. I realize you can get a few here in the US now, but you get the point. This is making your niche look like a regular niche is 98%...LOL. Your situation is a NICHE of the NICHE. Check steampowered survey if you don't get what I just said.
http://store.steampowered.com/hwsurvey/
Less than 1% run your res tested here. That's niche of a niche right? The entire group of people above 1920x1200 is less than 2% added all up (and this is out of probably over a few hundred MILLION steam users). Just click the monitor res and it will break them out for you. You wrote an article for NOBODY to show AMD doesn't suck vs Intel? Start writing for EVERYBODY (that other 99%) and you'll be making recommendations for INTEL ONLY.
I'm not saying anything bad against Ian here, clearly he did a lot of work. But whoever is pushing these articles instead of FCAT etc is driving this website into useless land. You guys didn't even mention NV's killer quarter (AGAIN). Profits up 29% over last year, heck everything was up even in a supposedly bad PC time (pc sales off 14%...no affect on Nvidia sales...LOL). They sell cards because their drivers don't suck and a new one comes out for every AAA title either before or on the day the game comes out! That's what AMD should be doing instead of console dev. They gave up the cpu race for consoles too! I'll be glad when this gen (not out yet) of consoles is DEAD. Maybe they will finally stop holding us back on PC's. They stuck us with 720p and dx9 for years, and they're set to stick us at 1080p for another 8yrs. They also allowed NV to not do anything to improve drivers for a year (due to AMD not catching them until Never Settle end of Nov2012). But maybe not this time...LOL. DIE CONSOLES DIE! :)
Here's what happens when you show 1080p with details down...cpu's part like the red sea:
http://www.tomshardware.com/reviews/neverwinter-pe...
Look at that separation!
"It's a little surprising that the Core i3-3220, FX-4170, and Phenom II X4 960 aren't able to manage a minimum of 30 FPS, though they come close. The dual-core chips are stuck at about 20 FPS, and the FX-8350 does a bit better with a 31 FPS floor that averages closer to 41 FPS. Only Intel's Core i5-3550 demonstrates a significantly better result, and we have to assume that higher-end Core processors are really what it takes to let AMD's single-GPU flagship achieve its best showing."
Note only two CPU's managed above 30fps minimum! I guess you need a good cpu for more than just CIV 5 huh? You should have ran at this res with details down to show how bad AMD is currently. PEOPLE, listen to me now. Buy AMD cpus only if you're REALLY taxed in the wallet and can't afford INTEL! I love AMD, but if you value your gaming fun (meaning above 30fps) and have a decent card, for the love of god, BUY INTEL. This was a test with a SINGLE 7970ghz. AMD is light years away from Taxing their won top end gpus. But Intel isn't. The bottom to top in this article at toms was 17fps to 41fps. THAT IS HUGE! And they didn't even show top i7's. It would likely go up into the 50's or 60's then.
Anandtech (not really blaming Ian himself here) is steering people into stupid decisions and hiding AMD's weaknesses in cpu's here, and in FCAT/gpu's with Ryan. I can't believe I'm saying this, but Tomshardware is actually becoming better than anandtech...LOL. WOW, I said that out loud. I never thought that would happen. It's 4:50am so I'm not going to grammar/spellcheck the nazi's can have fun if desired. :) Too bad I got to this article a week late.
http://techreport.com/review/23246/inside-the-seco...
THE REAL CPU ARTICLE YOU SHOULD READ. Note the separation from top to bottom in skyrim here is 58fps for AMD up to 108fps for Intel...See my point? Leave it to Scott Wasson (the guy who broke out the need for FCAT! along with Ryan Shrout I guess at pcper) to write the REAL article on why you don't want a slow cpu for ANY game. This is what happens at 1080P people! Note the FX8350 and 1100T are nowhere NEAR Intel in this review in ANY game tested. The phenom ii x4 980 is slow as molasses also! Note also Scott discusses frametimes which show AMD sucks. Welcome to stutter that isn't just because of the gpu...LOL.
" All of them remain slower than the Intel chips from two generations back, though. "
And this one sums it up best on the conclusion at techreport's article:
"We don't like pointing out AMD's struggles any more than many of you like reading about them. It's worth reiterating here that the FX processors aren't hopeless for gaming—they just perform similarly to mid-range Intel processors from two generations ago. If you want competence, they may suffice, but if you desire glassy smooth frame delivery, you'd best look elsewhere. Our sense is that AMD desperately needs to improve its per-thread performance—through IPC gains, higher clock speeds, or both—before they'll have a truly desirable CPU to offer PC gamers. "
Only anandtech has AMD rose colored glasses people. READ ELSEWHERE for real reporting. So AMD doesn't even offer a desirable cpu for gamers...LOL. Sad but true. Toms shows it, techreport shows it and if I had more time people, I really rip these guys apart at anandtech by posting a few more cpu tell-alls. This site keeps putting up stuff that HIDES AMD's deficiencies. I'd like to buy an AMD cpu this round, but I'd be an idiot if I did as a gamer. I7-4770k for me sir. Spend whatever you can on a haswell based system (it supposedly takes broadwell later) and wait for 20nm gpus until xmas/q1 where the real gain will come (even low end should get a huge bump). Haswell comes next month, you can wait for the FUTUREproof (if there is such a thing) socket one more month. Trust me. You'll be happy :)
I'd put more links, but this site will see too many and call me a spammer...UGH.
colonelclaw - Monday, May 13, 2013 - link
You lost me at '...same crap Ryan...'Never a great idea to preface a wall of text with an insult.
TheJian - Tuesday, May 14, 2013 - link
Well they have previously done worse to me :) I presented data in the 660ti article, called out their obvious lies even with their own data (LOTs of Ryan's own benchmarks were used to show the lies), which prompted Jarred to call me a Ahole and said my opinion was UNINFORMED ;). Ryan was claiming his article wasn't for above 1920x1080 (or 1200) but he was pitching me $600 Korean monitors (same ones mentioned here) you had to buy from EBAY and give you Visa to a nobody in Korea. Seriously? It could not even be bought on amazon from anyone with more than a SINGLE review, which I pointed out was probably the guy reviewing himself :) He had no about page on his site, no support etc, not even a phone#, just an email if memory serves. It was laughable. After taking Ryan down, Jarred attacked ME not the data.What do you expect a person to do after that?
They've been snowing people for a long time with articles like this.
Where is FCAT article part2? Where is the FCAT results from 7990? We are still waiting for both and will continue as I keep saying until AMD fixes their junk drivers and I guess gives a green-light for Ryan to finally write about FCAT for REAL. This is a pro AMD site (used to be more neutral!), I really didn't write it hoping to get love from the viewers. I just wanted the data correctly presented which other sites did with aplomb. You don't have to like me, or the data, just realize it makes sense as shown in the post via links to others saying it. NOT me.
People who stopped at "same crap ryan" were not my intended audience ;) I can hate a person (well I never do) and still value the data in a great argument made by said person. I don't care about them as long as it makes sense. The person means nothing. As I said above I don't blame IAN really, he's just doing what he's told. I even admired the work he put in it. I just wish that work could have been dedicated to data actually useful to 98% of us instead of nobody while hiding AMD's weaknesses. AMD is not a cpu I could recommend currently at all for anything unless you are totally strapped for cash. Even then, I'd say save for another month or something and come home with Intel. I'm not really an Intel fan either...LOL. I was selling AMD back when Asus was leaving their name off their boards (fear of Intel) and putting their motherboards in WHITE boxes! Intel should have had to pay AMD 15B (they made 60+B over the years screwing AMD like this). They had the best cpu's and Intel managed to stall with nasty tactics until they caught them. I love some of Intel's chips but generally hate the company. But I'd consider myself a D-Bag for not telling people the truth and shafting their computer purchase doing it. If I personally want to help AMD and buy a chip I think is way behind, great - I've done my part (I won't just saying). But I wouldn't tell my friends to do it claiming AMD is great right now. That's not a friend. Anandtech's readers are in a sense their friends (we keep reading or they go out of business right?). Hiding things on a massive scale here is not what friends do for friends is it?
I didn't expect any favorable comments from consoles lovers either :)
OwnedKThxBye - Tuesday, May 14, 2013 - link
We might all hate this guy (for good reason) but the words he writes regarding CPU performance in this article have a lot of truth.yhselp - Tuesday, May 14, 2013 - link
Agreed. What he wrote is offending, emotional and hardly objective. However, there's a truth hidden in there somewhere. Consider the following scenario.Here are a few suggestions. Since most users that would spend $500 on a flagship video card and $600-$800 on a 1440p monitor and God knows how much more on the rest of the system, aren’t likely to skimp on CPU choice to save a hundred bucks, a different testing scenario might produce more useful information for the masses (regarding cheap/er CPUs for gaming).
A more likely market for an AMD CPUs in a gaming rig would be people on a tight budget – when every buck matters and the emphasis is on getting as fast a GPU as possible. In my opinion, it’d be quite useful to test various AMD CPUs which are cheaper than an Intel quad-core; paired with a 650 Ti Boost and/or 600 and/or similarly-priced AMD video card at 1080p. Of course, this would raise yet another question – are Intel dual-cores faster than similarly-priced AMD quad-cores in this mid-range gaming scenario?
Suggestions for other CPUs:
Core i5-3350P – baseline Intel quad-core performance (cheapest Intel quad-core on the market)
Pentium G2120 – should perform similarly as an i3 for gaming (costs less)
Celeron G1610 – cheapest Intel CPU