AMD FX-8320E CPU Review: The Other 95W Vishera
by Ian Cutress on January 13, 2015 10:00 AM ESTProfessional Performance: Windows
Agisoft Photoscan – 2D to 3D Image Manipulation: link
Agisoft Photoscan creates 3D models from 2D images, a process which is very computationally expensive. The algorithm is split into four distinct phases, and different phases of the model reconstruction require either fast memory, fast IPC, more cores, or even OpenCL compute devices to hand. Agisoft supplied us with a special version of the software to script the process, where we take 50 images of a stately home and convert it into a medium quality model. This benchmark typically takes around 15-20 minutes on a high end PC on the CPU alone, with GPUs reducing the time.
A highly parallel benchmark like Photoscan puts the FX CPUs ahead of our i3 results. When overclocked, it plays just behind the i5 results.
Cinebench R15
Professional Performance: Linux
Built around several freely available benchmarks for Linux, Linux-Bench is a project spearheaded by Patrick at ServeTheHome to streamline about a dozen of these tests in a single neat package run via a set of three commands using an Ubuntu 14.04 LiveCD. These tests include fluid dynamics used by NASA, ray-tracing, molecular modeling, and a scalable data structure server for web deployments. We run Linux-Bench and have chosen to report a select few of the tests that rely on CPU and DRAM speed.
C-Ray: link
C-Ray is a simple ray-tracing program that focuses almost exclusively on processor performance rather than DRAM access. The test in Linux-Bench renders a heavy complex scene offering a large scalable scenario.
Ahead of the i3, behind the i5 until overclocked seems to be a good motto for this CPU.
NAMD, Scalable Molecular Dynamics: link
Developed by the Theoretical and Computational Biophysics Group at the University of Illinois at Urbana-Champaign, NAMD is a set of parallel molecular dynamics codes for extreme parallelization up to and beyond 200,000 cores. The reference paper detailing NAMD has over 4000 citations, and our testing runs a small simulation where the calculation steps per unit time is the output vector.
NPB, Fluid Dynamics: link
Aside from LINPACK, there are many other ways to benchmark supercomputers in terms of how effective they are for various types of mathematical processes. The NAS Parallel Benchmarks (NPB) are a set of small programs originally designed for NASA to test their supercomputers in terms of fluid dynamics simulations, useful for airflow reactions and design.
Redis: link
Many of the online applications rely on key-value caches and data structure servers to operate. Redis is an open-source, scalable web technology with a b developer base, but also relies heavily on memory bandwidth as well as CPU performance.
It would seem that one downside to the AMD FX line is the memory access limited scenarios, unless more cores can come into the equation. For redis, that is unfortunately not the case.
Facebook
Twitter
RSS
92 Comments
View All Comments
LeptonX - Tuesday, January 13, 2015 - link
How long do you test with OCCT and what data size? I ask because I often have crashes after 3-6 hours and if I want total stability that lowers my OC by as much as 200MHz.Oxford Guy - Sunday, April 19, 2015 - link
If the chip even lasts that long at an 1.550 volts and what is most likely a temperature well above the rated maximum.bsim500 - Wednesday, January 14, 2015 - link
"Idle to Delta Power Consumption":-"95w" FX-8320E = 86w
"95w" FX-8370E = 127w
"125w" FX-8350 = 163w
"220w" FX-9590 = 272w
And this is precisely why people want the actual idle & load figures not worthless "delta" scores that still do not reflect if, eg, one platform is idling 20w higher than another, etc. It doesn't involve any extra work since you need to acquire both anyway to calculate the delta!
yannigr2 - Wednesday, January 14, 2015 - link
Every time I see charts like these in the article, I am hitting my head on the wall shouting"32nm Thuban. 32nm Thuban you morons!".
Cryio - Wednesday, January 14, 2015 - link
Anandtech guys, please, sometime in 2015 please update the charts with some newer games, that are also CPU dependent.• Tomb Raider is exclusively GPU dependent.
• F1 2013 is old, 2014 was released some time ago. You should still probably benchmark GRID: AutoSport, since it's the newer game with the better optimized engine. F1 games always done poorly with AMD CPUs, for whatever reason.
• Battlefield 4 should be test only on MP. And if you really want to enphasize that CPUs can do in that game, get a HD 7000/R 200 series AMD GPU and run the game on Mantle to see what the CPU can do.
• Get an RTS game on that list.
• Get Far Cry 4, Watch_Dogs, Crysis 3 or Metro Last Light Redux on this list. These are properly CPU hungy games.
Paddockrj - Wednesday, January 14, 2015 - link
Dear members of this channel. You are eating a bull but a fly make you all bleed! AMD created those processors because a lot of people that has Am3+ 95w mobo's need it! If you have a 8120...8150...8320...8350...8370 or FX6xxx, you dont need 8320E or 8370E! AMD created those processor for a good upgrade ONLY.In gaming, FX8320E with a gtx770 or r9 280x gets the same fps that any Intel Core i. Some fps more, some fps less, but the same level. The diference happens when we try to play a Intel optmized game or a game that uses 2 or 4 cores better... MUCH BETTER. 1 core of Intel is better than 1 core of AMD, but we must see that in multicore, FX is excellent. Only i7 2nd or better can face a FX8350 per example in multicore.
But now, the most important perfomance is single core. And this way, Core i3 and i5 4th perform better. However it is not the FX end. I think within two or less years, when 4k be a truth here in Brasil, who have a FX6xxx or better will give thanks God for that. Because 4K uses multicore too much and, I told you all, FX is excellent this track.
We dont need be fanboys, we are getting nothing. I like AMD cause my Phenom 2 x6 + r9 270 run games like an i5 and I pay R$ 1200,00 less at least... More or less = U$ 500,00. That mean, I pay less and I do the same thing. Sure, i5 can convert a film in 30s, my old X6 can do it in 45s, but when I try to convert 2...3...4 films at the same time, X6 can do it in 3minutes, i5 = 5minutes. I lost in single, but in multicore I win. That is matter to me.
All processors has you reason to exist! When people ask me: Ok, multicore is very good, but for gaming? AMD? Why?
I can answer using this example. A lot of people say: I will buy an i5 4430 and a gtx750... or r7 250 or 7750... or 7730... MAN! Why? Oh God! Why?
You must see that NOTHING... Read it well: NO-THING is more important than graphics card in gaming. If you will use a 7870 or r9 270 or gtx760 and want use this build over and over and over and over... Why dont you buy an FX6300 + mobo withi usb3, 1600mhz memory support? = R$650... But no! Intel is VERY VERY VERY HIGH MUCH ULTRA MAXED MAJOR... They think this way. Thinking this way, they buy an i3 4150 + mobo B85 = R$707...
Ok... "tell me the diferences, man!"
i3 will win fx6300 in single core test...
i3 will convert film faster if the program use 2 cores (rare!)
i3 will have 5fps more in some cases
i3 will lose in multicore test
i3 will suffer in 4K
i3 will cost more
And now I ask you: Why pay R$60 plus to do the same thing? With the same quality?
I cant find a usual answer...
JumpingJack - Thursday, January 15, 2015 - link
So what you are saying is that AMD's higher end desktop processors are essentially competitive with Intel low end desktop processors. No wonder they are cheaper.BrokenCrayons - Thursday, January 15, 2015 - link
I consulted my fortune cookie and zodiac to find out that delta charts for power are still pointlessly useless.corsa - Friday, January 16, 2015 - link
PCMark8 v2 OpenCLThe only Benchmark that has no scores for the Blue Team, were they to embarrassing to publish Ian?
Oscarcharliezulu - Sunday, January 18, 2015 - link
What's up with agisoft? Your link http://www.agisoft.ru shows an info box page saying they haven't plaid their hosting bills.