Archive Home arrow Reviews: arrow Processors arrow AMD FX-8350 Vishera Desktop Processor
AMD FX-8350 Vishera Desktop Processor E-mail
Reviews - Featured Reviews: Processors
Written by David Ramsey   
Monday, 22 October 2012
Table of Contents: Page Index
AMD FX-8350 Vishera Desktop Processor
Features and Specifications
Processor Testing Methodology
AIDA64 Extreme Edition Tests
PCMark 7 Pro Tests
CINEBENCH R11.5 Benchmarks
PassMark PerformanceTest
Media Encoding Benchmarks
SPECviewperf 11 Tests
SPECapc Lightwave
Blender
Piledriver Core Performance
AMD FX-8350 Overclocking
Piledriver Final Thoughts
AMD FX-8350 Conclusion

Handbrake Media Encoding

It's a truism that consumer-level computer performance reached the "fast enough" point years ago, where increases in system performance don't make thing any faster for most people. Web browsing, e-mail, word processing, and even most games won't benefit dramatically from a super-fast CPU. There are some exceptions, though, and media encoding is one of them: transcoding video, especially high-definition video, can bring the strongest system to its knees. Fortunately, media transcoding is one of those things that (depending on the design of the code, of course) that scales really well with both clock speed and the number of cores, so the more you have of both, the better your results will be.

The free and open-source Handbrake 0.96 video transcoder is an example of a program that makes full use of the computational resources available-- in other words, it'll keep all available cores saturated. For this test I used Handbrake 0.95 to transcode a standard-definition episode of Family Guy to the "iPhone & iPod Touch" presets, and recorded the total time (in seconds) it took to transcode the video.

handbrake.png

Video transcoding is one of the things that really makes good use of a multi-core processor. Thus the 8-core FX CPUs turn in better results than the 4-core 2500, with the stock-clocked FX-8350 being 25% faster than its Intel competition.

x264 HD Benchmark 5.0

Tech ARP's x264 HD Benchmark comprises the Avisynth video scripting engine, an x264 encoder, a sample 1080P video file, and a script file that actually runs the benchmark. The script invokes four two-pass encoding runs and reports the average frames per second encoded as a result. The script file is a simple batch file, so you could edit the encoding parameters if you were interested, although your results wouldn't then be comparable to others.

x64hd_5.0.png

Video transcoding applications show the AMD FX-8350 processor to its best advantage, keeping all eight cores busy. Still, it's telling that at stock clock speeds the 8350 is only marginally faster than the four-core 2500K, which is also running at a lower clock speed.



 

Comments 

 
# RE: AMD FX-8350 Vishera Desktop ProcessorAthlonite 2012-10-22 21:35
Can I ask you did you install the 2 hotfix updates for windows 7 for the FX CPU's
the threading and core parking updates can make a bit of a difference to benchmarks especially in lightly threaded situations
Report Comment
 
 
# RE: RE: AMD FX-8350 Vishera Desktop ProcessorDavid Ramsey 2012-10-22 21:48
Yes, the hotfixes were installed, but as tests on other sites have noted, they make very little difference.
Report Comment
 
 
# Tier 2 Tech Agent.John Andersen 2012-10-22 23:26
I have an Intel I7-2700K@4500Mhz on a Corsair H100 lowest setting. - Seems like I can keep that a while longer.
Report Comment
 
 
# fx6100@4.4ghz to fx8350?major 3000 2012-10-23 04:03
I already own a 1st generation fx chip that I got for $77.00 and was wondering if this upgrade would be worth it when compared to the $ I paid.
Report Comment
 
 
# RE: fx6100@4.4ghz to fx8350?David Ramsey 2012-10-23 07:26
No, it wouldn't be worth it.
Report Comment
 
 
# Upgrad for less,.......RealNeil 2012-10-23 07:46
I really like that AMD is sticking to the same socket with these new CPUs. My path to an upgrade is a little less expensive because of it, and I can appreciate that on a retirement income.

The extra performance is not Earth shattering, but it is good enough to consider buying into. I will not be selling my i7-2600k system anytime soon, but I can see an upgrade to this new FX-8350 happening within a few months.
Thanks for the review David.
Report Comment
 
 
# You must test on Windows 8Kirk Martin 2012-10-23 09:59
If it is possible I would like to see the tests done on Windows 8, in a few weeks time, as we have been told that the AMD FX series processors have been designed for Windows 8. I would like to see if there is a major difference...
Report Comment
 
 
# RE: You must test on Windows 8David Ramsey 2012-10-23 12:59
Yeah, well, there's a problem testing with Windows 8.

When we review motherboards, processors, and the like, test systems are built, a fresh copy of Win 7 is installed, and we're good to go. Since Windows 7 will run for 30 days without requiring an activation code, it's easy to create fresh installations as needed. After the review is done, test systems are generally torn back down and the parts saved for future tests/comparisons.

Windows 8 requires a valid activation code just to install it-- you can't skip this step as you can with Win 7. Then it's locked down to the system it was installed on.

Right now that means we'd have to buy a new copy of Windows 8 for every video card, motherboard, or CPU we test. This isn't viable, and we're trying to figure out what to do about it.
Report Comment
 
 
# windows 8 consumer preview testingmajor3000 2012-10-23 13:24
That is not entirely true because you can get a windows 8 x86 or x64 consumer preview copy directly from Microsoft for free! It has a open serial key for people who want to try it... you could just test the FX 8350 on that please.
Report Comment
 
 
# RE: windows 8 consumer preview testingDavid Ramsey 2012-10-23 21:37
The problem with that is that the Consumer Preview isn't the same as the release version. In general we try to avoid testing with pre-release or beta software.
Report Comment
 
 
# RE: RE: windows 8 consumer preview testingnobody 2012-10-25 04:18
MSDN subscription is your best solution
Report Comment
 
 
# RE: RE: RE: windows 8 consumer preview testingAustin Downing 2012-10-25 04:55
Would you like to pay for that? Because if you will I will gladly use Windows 8 in all of my reviews.
Report Comment
 
 
# RE: RE: RE: windows 8 consumer preview testingDavid Ramsey 2012-10-25 07:44
I have a Microsoft Technet subscription. It gives me 3 license codes for each version of Wndows. (Used to be 10, then 5, now 3). So I could build three test systems with Windows 8 basic edition, assuming I didn't want to use any of them for myself. Throw in Windows. 8 Pro and I could build 6!
Report Comment
 
 
# no one cares about windows 8sam 2013-04-27 10:59
Why test on an OS that no one uses?
Report Comment
 
 
# amd STRIKES BACKbOLk 2012-10-23 14:15
Well, the introduction of these processors will give a good fight to Intel, and is what many expected from AMD.
Report Comment
 
 
# test on windows 8 consumer preview!major3000 2012-10-23 14:38
I would like to see this tested on Microsoft windows 8 consumer preview! It should not be a problem obtaining a copy since its free to download from Microsoft web site.
Report Comment
 
 
# Not accurateUserofPC 2012-10-24 04:20
these benchmarks don't match up to other benchmarks found on the web. Tom's show a better bench than this. I believe you have bad machine for testbed or you jacked the results in favor of Intel
Report Comment
 
 
# RE: Not accurateAustin Downing 2012-10-24 05:03
Beyond any other holes in that argument. Let me ask you this. What incentive would we have to skew the results?
Report Comment
 
 
# RE: Not accurateDavid Ramsey 2012-10-24 08:32
Really? How do you figure that Tom's "show a better bench" than we do?

'cause I just looked at their review and they used a different motherboard, different memory, different hard drive, and different video card, as well as an almost completely different set of benchmarks.

The only two benchmarks we have in common are PCMark 7 Pro and Handbrake. You can't compare the Handbrake results since we used different input files with different settings.

PCMark 7 Pro tests the entire system, not just the processor (as I specifically mention in my review), so it's not surprising that with different hardware they get different results. Our Creativity scores were virtually equal (less than 0.5% difference); our Computation score was much better (6468 vs. 5083) and our Productivity score was less.

Which proves precisely nothing, except that you have no idea whatsoever what you're talking about.
Report Comment
 
 
# RE: Not accurateOlin Coles 2012-10-24 08:46
@
it's silly for you to make claims without a single modicum of proof to support the argument. As the author points out, you can't compare completely different computer systems and expect the same benchmark score. Perhaps these in-depth technology reviews require a level of understanding you lack.
Report Comment
 
 
# 8150 AND SCORESMARIO OSUNA 2012-10-24 18:01
CHECK MI SCORE, AND OVER INTERNET ARE LIYING ABOUT AMD. I KIK SOMES I5, INTEL FAN BOY WANTS TO KEEP AWAY FROM AMD. AMD ALL THE WAY ...#imageshack.us/a/img89/3456/scoresh.jpg

SPECS.DDR3 GSKILL 2133MHZ.. PHENOM 3.9-4.2TURBO. 3 SSDS-RAID0
Report Comment
 
 
# RE: 8150 AND SCORESOlin Coles 2012-10-24 18:16
Find your CAP LOCKS key, then learn how it works.
Report Comment
 
 
# RE: RE: 8150 AND SCORESAustin Downing 2012-10-25 09:18
Don't you know? Caps lock is cruise control for cool.
Report Comment
 
 
# RE: 8150 AND SCORESdavid 2012-11-18 16:06
I will put my score on my 3930 6core intel up against any amd out there you will be shamed to the CURB MARIO YEA INTEL ROCKS
Report Comment
 
 
# RE: RE: 8150 AND SCOREShotshot 2012-12-02 00:34
you can point to a graph and show me your system wins, I point to my wallet and say I am the winner.Nowadays everyone gets so caught up comparing numbers on a chart instead of actually using the systems.I hope we never go back to the days when Intel is the only choice.
Report Comment
 
 
# RE: RE: RE: 8150 AND SCORESDavid Ramsey 2012-12-02 08:11
Man, I wish more readers had your attitude!
Report Comment
 
 
# RE: RE: RE: 8150 AND SCORESsam 2013-04-27 11:13
Wanting to use the CPU is the whole reason that we would buy a CPU that costs 3 times more. This thing while overclocked can barely beat an i5 that only costs about $30 more (in fact, it doesn't beat it in many tests). That's 8 cores that can barely beat 4 cores. It get's completely stomped by the i5 in any single core test which would make it completely useless for gamers. For business I would gladly pay 3 times the price for a 6 core intel CPU because it would absolutey destroy the AMD CPU. If all you want to do is browse the web and brag about how cheap your CPU is, then buy this. If you want to get stuff done, buy intel. I used to use AMD cpus for many years, but they just can't compare to what an intel CPU can do, even when they have twice as many cores.
Report Comment
 
 
# Variations around Webk brian 2012-10-24 21:44
There seems to be some differences between reviews that I have seen. I guess due to differences in which benchmarks are being run (amongst other variables) there have some different interpretations. Some I have seen note little differences between PD and BD, and others note advantages PD has over i5- albeit with much higher power usage.
Report Comment
 
 
# RE: Variations around WebDavid Ramsey 2012-10-24 21:51
Exactly. Each site has their own hardware and testing regimen. In general results for things like CPU and video card tests will be similar across sites, but a lot depends on how the benchmarking is done.

My personal philosophy for CPU reviews is to run a LOT of benchmarks, since different benchmarks show different things.

The reviews I've read on other sites so far seem to have all reached similar conclusions: faster than Piledriver, competitive in most uses with midrange Intel CPUs.
Report Comment
 
 
# RE: AMD FX-8350 Vishera Desktop ProcessorMatt W 2012-10-31 10:33
Just wanted to say thanks for including Lightwave rendering benchmarks. Finding the most rendering power for the money is really helpful to me. While Intel may have faster chips, AMD's lower cost CPU's and motherboards give me the best bang for the buck.
Report Comment
 
 
# DRAM Testing w/ 8350James Reece 2012-10-31 14:43
Wondering if any additional testing was done DRAM wise. I remember they also claimed initially that the 8150 ran 1866 DRAM natively, and finally retracted that to show in their advertising that it could run UP TO 1866 at 1 DIMM per channel...Any testing done with 4x4GB or 4x8GB at 1866 or higher. I ask because if multitasking is where this shines in today's world, the CPU needs to be able to run DRAM to support the multiple programs, especially if getting into multiple VMs.
Report Comment
 
 
# RE: DRAM Testing w/ 8350Olin Coles 2012-10-31 14:49
That wouldn't be a part of any CPU review, since it's more of a motherboard feature/limitation. The CPU can operate with any frequency, but the motherboard needs to support the speed and density first.
Report Comment
 
 
# RE: AMD FX-8350 Vishera Desktop ProcessorJames Reece 2012-10-31 19:08
I beg to differ, without going to extremes, CPUs do have DRAM limitations and it's not based on motherboard features/limitations...A typical Z77 motherboard paired with a decent 3770K (like anything else some are better than others) can run 32GB of 2600-2800 DRAM or better and without drastic cooling and run as a stable system. Take the same DRAM and a 3570K and it's not going to happen, the 3570K typically can run up to 2400 in a 32GB setup (again with a decent 3570K as many can't run 2400), put a 2500K in the same motherboard and now your down to about 2133-2200.

The 8150 came out claiming native 1866 support, which to most implies one can fill the DRAM slots and run 1866, after untold complaints from people trying to run just 4x4GB (16GB total) of 1866, the advertising was changed to reflect 1866 at 1 DIMM Per Channel. AMD's BIOS and Kernel Guide for Developers reflects the same info (and that was based on 4GB DIMMs)....

I ask about the DRAM as AMD has been notorious for having weak Memory Controllers, (i.e. the initial 965 in general, couldn't carry 1600 DRAM until the C3 revision), so the CPU itself, primarily it's Memory Controller, does have a major say in what DRAM (frequency, density, total) it can run.

I, and I know of many others, would love to give this CPU a try, and see how it would do with the 8 cores....if it can support a sufficient amount of DRAM of a high frequency to run multiple VMs and applications. If in fact it runs 1866 as 'native' as AMD claims would be nice to see it run 32GB of same and if it CAN run faster DRAM.

Also of note: Would be interesting to see how the 8350 stands straight up to a 2500K with both OCed - most 2500Ks will run easily stable at 4.7 or better - vs the 5.0 of the 8350, keeping in mind of course the 2500K is now almost 2 years old, or better yet vs the 3570K (generally will run 4.5-4.6 with few problems or heat), which as of today at both the Egg and Tiger is about $5 less than the 8350
Report Comment
 
 
# RE: RE: AMD FX-8350 Vishera Desktop ProcessorDavid Ramsey 2012-10-31 19:27
I'm not sure why you're so concerned with memory frequency. The CPU and ASUS Crosshair V Formula motherboard handles DDR3-1600 4x4GB just fine (that's what I've got in it now), and you'd be hard pressed to show any difference in performance with DDR3-1866 in anything except a memory benchmark.
Report Comment
 
 
# RE: RE: AMD FX-8350 Vishera Desktop ProcessorDavid Ramsey 2012-10-31 19:31
Oh, yeah: we don't normally compared multiple overclocked processors simply because no overclock is guaranteed: we reached 5GHz on the 8350, but several other review sites did not. Throwing another overclocked CPU into the mix just reduces the chances that anyone would be able to duplicate our results. That said, I'm pretty sure an overclocked 2500K would be an overclocked 8350 in everything except highly threaded workloads, considering how close the stock-clocked 2500 scores are. As for the 3570K, we've never received one to review and so don't have it available. FWIW, I just checked at Newegg, and the 8350 is $10 less than the 3750K...or it would be had they any in stock.
Report Comment
 
 
# RE: AMD FX-8350 Vishera Desktop ProcessorJames Reece 2012-10-31 20:09
After the $229 price, note the promo code for an additional $15 which ends tomorrow, was originally through the 6th of Nov as of this AM, but has been cut back based on sales being so high, Tiger Direct has them at $214.99 after coupon (same/same, depends on sales how long it last.

Newegg "$15 off w/ promo code EMCJNHD27, ends 11/1"

##newegg.com/Product/Product.aspx?Item=N82E16819116504&Tpk=3570K

As far as memory frequency if there's nothing to be concerned about between 1600 and 1866, then why not test at the CPU's true 'native' freq of 1333? Regardless, my question was how about large amounts of DRAM at high freq, i.e. can it handle 32GB @ 1600, 1866, 2133, 2400 etc. I originally brought up the question as the testing was done with 8GB, not the 16 you have loaded currently. I know most of my customers, look for 16GB to start and even systems that are planned primarily for gaming, those folks want 8GB min to start....and when looking to enthusiast systems, most want to get the most out of the system they can, whether it be gaming, rendering video, imaging work, GIS with large data sets, VMs where higher freqs = greater MTs (Mega Transfers per sec) or higher bandwidth. Maybe it's just me, but when I pull my DRAM out of one of my systems 2133 or 2400 and put in 1600 or even 1866 and run that in a like amount, I can 'see' a difference, especially when I start opening up say Win8 in a VM and play with images and be working with GIS or video in Win7
Report Comment
 
 
# RE: RE: AMD FX-8350 Vishera Desktop ProcessorDavid Ramsey 2012-10-31 20:28
I test at DDR3-1600 because that's the most common memory frequency these days, and it's what most readers are likely to have. Sadly I do not have any 32GB memory kits available...one thing you may not realize is that we do not have a giant storeroom full of hardware to draw on; all we can test with is the hardware we have available. Your customers' needs-- 32GB systems running multiple VMs-- are not those of our typical readers, either.

Here at Benchmark Reviews we have had occasion to test high frequency memory, and we've never found any quantifiable real-world difference in performance other than on synthetic memory benchmarks. If you can suggest a real-world application that benefits, with some numbers to back it up (rather than your 'feel'), we'd be interested to here about it.
Report Comment
 
 
# RE: RE: AMD FX-8350 Vishera Desktop ProcessorDavid Ramsey 2012-10-31 20:33
Oh, yeah: check the memory reviews section here for my review of Kingston HyperX DDR3-2133 memory. The only application I could find that benefited noticeably from the higher speed memory was Euler 3D, and the difference between DDR3-1333 and DDR3-2133 was a whopping 11%. The other application tests showed no gain at all. Considering the very high price of the memory and the minimal to nonexistent performance benefits, it would be the very last place in my system I'd look to for performance increases. If you want your system to "feel" faster, get an SSD. No other single change will have a more noticeable effect on performance.
Report Comment
 
 
# RE: RE: RE: AMD FX-8350 Vishera Desktop ProcessorJames Reece 2012-10-31 22:31
Yes 1600 is the most ?common? as far as sets available, it?s basically entry level for Intel CPUs and mid to high for most AMD CPUs, but isn?t this about an enthusiast?s system/CPU? The ranks of available 1333 sets continues to shrink and even the quantity of available 1600 sets is going down as sets of 1866-2800 continue to increase, and with DDR4 around the corner look for these trends to increase in their scope.
I guess our thoughts as far as real world computing or quantifiable examples differ, I look to utilizing the cores of multiprocessor CPUs as an example, which is why I mentioned opening Win8 in a VM and rendering a video, while at the same time being in Win7 and running GIS, where as you talk about running a single app or benchmark at a time (i.e. how many of those 8 cores were actively doing anything in any one of the benchmarks, ever consider running multiple benchmarks at the same time?...might be surprised what you find). Further I said I could ?see? a difference when using faster DRAM, as in, things get done faster, not ?feel? as you imply somewhat sarcastically.
Report Comment
 
 
# RE: RE: RE: RE: AMD FX-8350 Vishera Desktop ProcessorDavid Ramsey 2012-11-01 07:05
I didn't mean to come across as sarcastic, but thirty years as a programmer has taught me that there's no substitute for actual testing, and the testing we've done here on high speed enthusiast level memory has never demonstrated more than a minuscule performance improvement on anything except synthetic memory bandwidth benchmarks. You can see the results of my own tests of a DDR3-2133 here: #benchmarkreviews.com/index.php?option=com_content&task=view&id=723&Itemid=67

Again, if you can suggest some real world tests that will show the advantage of high speed memory, we're all ears.
Report Comment
 
 
# RE: RE: RE: RE: RE: AMD FX-8350 Vishera Desktop ProcessorJames Reece 2012-11-01 07:50
Testing is true, I started programming in 1980 with mainframes, but to this day have found few benchmark programs that really delve into DRAM (other than reads/writes/copies), but the nice thing with DRAM, if in sufficient quantities, it remains constant in a true multitasking or multi user environment, unlike CPUs and GPUs which bog down as additional tasks are thrown at them. The DRAM continues you to feed at the same speed, so in effect the faster the better. As I mentioned if you try running multiple benchmarks at the same time, even with a quadcore or a 8 core you'll see dips in performance, or run a benchmark at the same time you're running a virus scan and have a few other programs open and look at the differences in results. Also look at the CPUs themselves, i.e. the MC (Memory controllers) on Intel are stronger than those AMD has so Intel handles and utilizes DRAM better, increasing overall performance. Rather than sit and run one benchmark after another try mixing things up and look at performance
Report Comment
 
 
# RE: RE: RE: RE: RE: RE: AMD FX-8350 Vishera Desktop ProcessorOlin Coles 2012-11-01 08:33
"Rather than sit and run one benchmark after another try mixing things up and look at performance"

Performance is NOT a warm fuzzy feeling you get, it's a quantifiable result. Anyone with an understanding of DDR3 memory will tell you that clock speed isn't important in contrast to latency. I even wrote a long in-depth technical article about: bit.ly/PoNdE8

Stop commenting, and start reading about the technology behind these items. You're asking us to run DDR3-2800 for the sake of a CPU test, which doesn't make sense to anyone who understands the pipeline architecture of a PC. Even if we wanted to run such a ridiculous test, finding compatible hardware components that consistently performed to this level would be extremely difficult.
Report Comment
 
 
# RE: RE: RE: RE: RE: RE: RE: AMD FX-8350 Vishera Desktop ProcessorJames Reece 2012-11-04 19:16
Again I'll beg to differ, you say clockspeed isn't important in relation to latency - Try running 8GB of CL9 1333 or 1600 vs CL9 2133 and then which would you prefer? The primary place where latency is more important is within testing sets rated for the same freq i.e. 2 sets of 1600, 1 at CL7 the other at a higher CL (8, 9, 10), the CL7 sticks should perform better. The FACT is and it is quantifiable, is that the true importance of latency and and clockspeed is a combination of the two, the preferred is the highest clockspeed/freq you can get at the lowest latency....As a matter of FACT, I believe in the article you mention YOU even stated that higher bandwidth can overcome latency or to that effect, and I'd have to find it, but in one of the reviews of DRAM (the 1866 Snipers) you basically agree with me saying it's a combination of both.

As far as the technology goes, I'm rather comfortable with my knowledge, which is one of the reasons I didn't ask you to test with 2800 as you claim, in fact for 2800 the 3770K is about the only CPU that can carry 2800, so it's ridiculous for you to claim I said that (yet both you and David seem to delight in twisting what people say here). I simply asked if faster sets were looked at, one reason being you all tested the 8150 with 1866 sticks, so why drop to 1600 when you have a newer more powerful CPU (and in fact the 8150 had better bench results running with 1866 that it does here running with more DRAM 8GB vs 4GB in the original test/review), yet only using 1600 here.

Also in reading the reviews here, as long as you brought up reading about the technology here, in the review on the 1866 16GB Ares, it states as a CON the DRAM XMP profile doesn't set the specified speed and timings, which as you should know, the DRAM itself doesn't actually set things, it provides information in the SPD and it is up to the BIOS to take in and implement that info (i.e. it's not the RAMs fault), in the same review it's said it's not as cost effective as buying two 8GB sets, which may be considered a con-but most DRAM manufacturers recommend against mixing two sets or more, even of the same exact model - the XMP programming is done by the set which will differ between 2 and 4 stick sets, often leaving one to have to manually set up the sticks (in particular the advanced/secondary timings like tRFC (which can be a bear for most all users). There's also other DRAM reviews that have questionable items
Report Comment
 
 
# RE: RE: RE: RE: RE: RE: RE: RE: AMD FX-8350 Vishera Desktop ProcessorOlin Coles 2012-11-05 09:34
I said "clock speed isn't important in contrast to latency", which is very different than "clockspeed isn't important in relation to latency". Perhaps you believe 'contrast' and 'relation' mean the same thing, but they don't. My point was fully explained in the article I linked.
Report Comment
 
 
# RE: RE: RE: RE: RE: RE: RE: RE: RE: AMD FX-8350 Vishera Desktop ProcessorJames Reece 2012-11-05 21:55
Could you elaborate as to what you actually mean and how it differs from what I explained...it's all fine and dandy to make far sweeping statements, but it would be nice if you explained them and provided examples as I did.
Another good example is: earlier you commented "That wouldn't be a part of any CPU review, since it's more of a motherboard feature/limitation. The CPU can operate with any frequency, but the motherboard needs to support the speed and density first"

Which I also differed with and explained, and you failed to reply to. According to that statement you imply an entry level socket 1155, I3 can run 32GB of 2800 DRAM...if the motherboard supports it...I disagreed and again explained myself, Could you also explain your premise to that statement? Just curious.
Report Comment
 
 
# RE: RE: RE: RE: RE: RE: RE: RE: RE: RE: AMD FX-8350 Vishera Desktop ProcessorDavid Ramsey 2012-11-06 07:40
Well, it's like this, James: no matter how you write a review, there will be people out there smugly certain that they could have done it better; that you overlooked an important facet; that your choice of benchmarks was wrong or that your review methodology was flawed, et cetera.

For example, as you point out, in my FX-8150 review I used DDR3-1866 memory on the AMD processor and DDR3-1600 memory on the Intel processor. The idea was to show each CPU in the best light with its highest officially supported memory frequency.

Wow, you should have seen the comments I got on that. For example, one commenter "Jim Reece" said "I'll give you that you did your best to cripple the 2500K", because I didn't run 1866 DRAM with it. He also took me to task for using a 2500K as the comparison (AMD's stated performance target) rather than a 2600K or 2700K, apparently under the impression that I have every Intel CPU ever made in a box here or something.

In the end this commenter was so disgusted with me, certain that I was simply a shill for AMD, that he said "No need to reply, I'm gone, I'll stick with sites that ACTUALLY perform reviews that are objective rather than w/ bias."

But you know what? He came back. Not only did he come back, he applied to be a reviewer here. After missing several promised dates to complete his first review, he informed us that he wouldn't be a "good fit" because he's a "straight shooter" (and, as he informed us a year or so ago, we're biased) and launched into a bizarre screed attacking several of our other reviews on technical points, all of which were wrong.

We hope he's gone for good this time.
Report Comment
 
 
# RE: RE: RE: AMD FX-8350 Vishera Desktop ProcessorJames Reece 2012-10-31 22:33
As far as ?getting? a SSD, it?s a good idea, I?ve been using them about since they came out, and when set up correctly they are excellent and will last a long time (by setting them up correctly, keep in mind that writes to the SSD are its bane, so you want to redirect as many writes as possible to a conventional mechanical drive or other device, this would include data files that are often changed, cache files, temp files, etc. To take the SSD suggestion one better, I?ve found for most people a 120GB SSd to suffice for most people for their OS and programs, generally for the same price or less, you can go with two 60-64GB SSDs and put them in RAID 0 (and with Win7 and 8 if set up using Intel?s RAID drivers as opposed to say a motherboard specific driver like Marvel or one of those, you can move the raid from one system to another rather seamlessly), this provides even more speed than a single drive.
Report Comment
 
 
# SSD IN RAIDRomano 2012-11-06 07:46
Thanks for your advice, never thought about putting 2 SSD drives in RAID. THANKS AGAIN
Report Comment
 
 
# RE: RE: RE: AMD FX-8350 Vishera Desktop ProcessorJames Reece 2012-10-31 22:34
Another thing you can do for a large jump in performance, is since the writes need to be redirected anyway to keep the health of the SSD up and prolong it?s life, if you have a large quantity of DRAM you can easily set up a RAM Drive and redirect all those writes to it which in particular can speed up browsing and other apps that create/use a large number of temp files (and further, with a healthy amount of DRAM to start with, you reduce the number of writes to disk (or RAM Drive anyway and much more of the data can be held in DRAM). Wish more room was allowed for comments
Report Comment
 
 
# A DISK MADE OF RAM?!???Farnsworth Worthington 2012-11-16 16:58
Wait, so you can set up a thing that is made of RAM and acts as a disk drive? So, this "RAM-DISK", as one might call it, is super fast like RAM but also able to act as a storage space like a traditional hard disk or SSD? WOW WOW I WILL GO TRY THIS AND SEE IF THIS "RAM-DISK" IS POSSIBLE.
Report Comment
 
 
# RE: A DISK MADE OF RAM?!???James Reece 2012-11-16 17:12
Yes, DataRam has a free version of their RAMDisk software, it goes up to 4GB and there are others out there, but might try this first, can find it here:

#memory.dataram.com/products-and-services/software/ramdisk
Report Comment
 
 
# MrPeter Stanich 2013-04-01 17:10
Windows 7 Calls it ReadyBoost and best used with a 16-64 GB Flash drive in 3.0 USB, very effective in SSD C: Drive Builds
You can select how much of drive you use while retaining programs on it.
Windows recommends using a Flash Drive that can double the size ofyour RAM minus any programs U want to keep onit
Report Comment
 
 
# Two different thingsJames Reece 2013-04-01 17:19
Ready Boost and a RamDisk are two entirely different things, Ready Boost is part of cache system and operates off a flash drive, SSD, or any portable flash storage device, a RamDisk uses the actual DRAM itself and is much. much, much fastertz7jt
Report Comment
 
 
# A8-5600 vs FX-4 4300mjssge 2013-02-05 10:46
Hi ;-)
Just been reading your benchmark test on the new trinity APU's (very good article), I used the A10 in a recent mini-itx build (Bit_fenix prodigy).

Ive had a request to build a rig for someone on a tightish budget, my first thought given the money available was to put in a AMD Piledriver FX-4 Quad Core 4300 and a 1gb 7850 gpu - total price: 210gbp, then I thought what about the A8-5600k crossfired with a HD6670 at approx: 134gbp.

I haven't been able to find any benchmarks unfortunately, I think its a pretty relevant question at the moment.

Many Thanks
Joel
Report Comment
 
 
# RE: AMD FX-8350 Vishera Desktop ProcessorNycto 2013-03-20 08:38
So I have been following an AMD upgrade path for awhile now. When my last motherboard died, an ASUS M2N32-SLI Deluxe, I bought a Sabertooth 990FX board as a replacement. I am still using the same Phenom II x4 980 CPU. I am currently very happy with my PC, however, I wonder, should I bother to update the CPU to something Piledriver based, or ride it out awhile longer? Are the advantages at this point worth the upgrade?

Thanks,
- Aaron
Report Comment
 
 
# RE: RE: AMD FX-8350 Vishera Desktop ProcessorDavid Ramsey 2013-03-20 08:59
Unless you run programs that benefit from heavy threading, like video encoding, then no: you won't notice any real performance benefit from the Bulldozer/Piledriver CPUs. Their individual core performance does not offer a significant improvement over the core performance in your Phenom II CPU (in some cases it's worse).

If you do run apps that can spawn a lot of threads, well, 8 cores rocks. But aside from transcoding and rendering, there aren't a lot of apps that can really leverage this power.
Report Comment
 
 
# piledrive theezjonny421 2013-03-21 12:53
thats really odd considering i just traded an fx4100 out for a 8350. the difference in gaming was pretty amazing. you can't compare core count on that. most games aren't even using 4
Report Comment
 
 
# 4x8GB DDR3 1600MHz or 2x4GBDDR3-1866?Alex 2013-05-30 18:10
Guys, I'm a bit confused here.

If I get a 2x4GB DDR3 1866MHz memory kit paired with the FX-8350 (working in dual-channel of course), will each module actually work at 1866MHz or what? Is it really much of an advantage to buy DDR3 1866MHz memory kits to be used on the AMD platform?

Thanks!
Report Comment
 
 
# RE: 4x8GB DDR3 1600MHz or 2x4GBDDR3-1866?David Ramsey 2013-05-30 18:39
Yes, each module will actually work at 1866MHz. With a dual channel setup, you'll be accessing two modules simultaneously for an effective doubling of memory bandwidth.

That said, in general high speed memory kits are more for racking up high benchmark scores than any noticeable improvement in performance. You'll be lucky to see an extra FPS or two. With 1600MHz memory being the standard these days-- and cheap as well-- there's little real-world reason to pay more for faster memory.
Report Comment
 

Comments have been disabled by the administrator.

Search Benchmark Reviews
QNAP Network Storage Servers

Follow Benchmark Reviews on FacebookReceive Tweets from Benchmark Reviews on Twitter