Performance test benchmark i7 n amd cores

leaf__00

Commendable
Oct 14, 2016
38
0
1,540
Well i x knw how to or where to begin, but try my best, will be long.
First of all english is not my native(gosh i its needless to mention but some prick love to do language critism,with intention to hurt or misundertood)

OK, now, Im building a budget gaming pc(not even 1080p) but just able to play any current or near future games(at medium settings only).I am using Performance test benchmark(i knw there r so many out but im used to this, n easier to make comparison between cpu n it works quite accurately).

i knw im noob, very noob, but try to understanf my "english" n speak my english because im not an IT student, srry.

!!!!!!PRIMARY QUEASTION: i want to learn which of the many tests( integer math,floating point math, prime number,multimedia instructions,compression,sorting, and single thread) relates or must be considered to judge a cpu's gaming performance.
For example, higher frequency always been offering gaming performance boost. But how aboout other.
For example, i7 2600 and fx 8350, both have almost same performance,, but there are test that i7 perform better than amd.For example encryption and compression test.Both are not related to gaming. Single thread performance is also better,but then without turboboost,as there are no games that runs on single cores, frequency per core should be poorer than AMD's.
AMD on the other hand have better score on some other tests.
Note that , i have been downloading many sample, with fair setups and with bench scores as consistent as possible for comparison.

So, does this dictates that AMD IS better for gaming.

Im seeing i5 forth generation having better score in cpu mark, however some subtest as very much to fx 6350.

So, pls to tell me what are test that i should care for or save to ignore( such as encryptions).

 
Solution

oobymach

Reputable
Jul 11, 2015
287
0
4,960
I game on amd so I have experience in this field, no it's not as fast as intel at single threading blah blah blah but in multithreading it is very close in performance with the 6700k at least according to cpu-z benchmarks.

My oc cpu vs 4ghz 6700k.
00cpuiz23oc_zpsenq1hqvg.png

I have the highest rank for an amd setup on catzilla 4k benchmark, I'm 46th on the list here, just a few points behind someone with an i5 and dual 1080's but mostly I'm ahead of all kinds of i7's with 3way sli's and titans and such.
http://www.catzilla.com/pl/toplist?page=3&res=2160&multi=all&ven=all&ctype=all&firm=all&max=460

So yeah, amd is alot better than some would like to admit.

Just because one benchmark shows intel leading doesn't mean this is always the case.

occt and prime 95 are good tools for testing overclocks on everything but the gpu (occt can test gpu but I wouldn't recommend it, it is more likely to fail than p95 during testing, I get about 1min50sec into occt before crash with my current oc but p95 passes lots of tests), for gpu testing get 3dmark or 3dmark11 to test your gpu and physx, catzilla is also really good and has a couple different tests like 3dmark to test physx and rendering and such up to 4k resolutions if you spend the $15 to unlock it like I did.

The gpu tests will fail if overclock is too high so you have something other than a game which, if games crash from high oc they can lose save data and other such problems.
 

CTurbo

Pizza Monster
Moderator
oobymach, your multi threaded benchmarks are irrelevant to gaming. Even today 98% of all games still only use one or two cores.

Even the $110 dual core i3 6100 is a better gaming cpu than the FX8350. Here is a review of an i3 4340 beating the FX8350 in almost (if not every) game. http://www.hardcoreware.net/intel-core-i3-4340-review/2/
 

oobymach

Reputable
Jul 11, 2015
287
0
4,960
Copied this from another thread as it's relevant.

Yeah but gta4's minimum requirement is a dual core amd or intel chip. It will 'run' on these chips but it was designed for a triple core cpu so dual cores lag horribly with it even though they meet the minimum required to 'run' the game.

Ever tried to play gta4 on a dual core? Even on a 3ghz athlon 64x2 I couldn't play with minimum settings without horrible lag. Minimum means it will barely run it.
Grand Theft Auto IV Minimum Requirements
CPU: Intel Core 2 Duo 1.8GHz, AMD Athlon X2 64 2.4GHz

Grand Theft Auto IV Recommended Requirements
CPU: Intel Core 2 Quad 2.4GHz, AMD Phenom X3 2.1GHz


You can play fallout 4 on a low end cpu but not without horrible lag, play it on an 8core at 4.6ghz it's smooth as silk.

The recommended requirements are recommended for a reason, the game was developed for them.
 
D

Deleted member 217926

Guest
D

Deleted member 217926

Guest

An i3 is a dual core with Hyperthreading. 2 cores 4 threads. ;)
 
D

Deleted member 217926

Guest
Oh and if you're really running SLI 1070s with an FX 8350 even one that's overclocked you are bottlenecking those cards. Put them with a stock 6700K and your FPS will jump 20% or better depending on the game engine.
 

bignastyid

Titan
Moderator
Funny you mention fallout 4
I3 6100 and a 980ti is better than a 9590 and a 980Ti.
http://www.techspot.com/review/1089-fallout-4-benchmarks/page5.html

In GTA V the older i3 4130 is only 6 fps slower than the 9590 and the I3 6100 is a fair amount faster than the 4130 so the 6100 will likely meat or exceed the FX.
http://www.techspot.com/review/991-gta-5-pc-benchmarks/page6.html

Plus the i3 6100 uses less power, produces less heat, costs less and can be used on cheaper motherboards without fear of VRM throttling.
 

Rogue Leader

It's a trap!
Moderator


To answer your question, correctly, the truth is there is no single metric that measures one better than the other. It all depends on how games are programmed and what type of game. The best thing to do is use multiple benchmarks to get a profile together of the processor. You will find (as linked above) current and even previous generation Intel processors easily demolish AMD in any gaming and most other real life tests.

In addition while you may hear that an FX-8350 can be heavily overclocked to be more competitive, it will cost you quite a bit more, between an expensive cooler, and a better quality motherboard you will spend nearly twice as much to get less performance than a stock i5-6500 on a $50 motherboard using the free cooler that comes in the box.

For example heres the benchmarks (and this site is good for setting up different comparisons):

http://www.anandtech.com/bench/product/697?vs=1646

And cost example (I know you are not in the US, this is just an example):

Intel:
PCPartPicker part list / Price breakdown by merchant

CPU: Intel Core i5-6500 3.2GHz Quad-Core Processor ($196.99 @ SuperBiiz)
Motherboard: MSI B150M Pro-VD Micro ATX LGA1151 Motherboard ($56.98 @ Newegg)
Memory: Kingston HyperX Fury Black 8GB (2 x 4GB) DDR4-2133 Memory ($41.99 @ B&H)
Total: $295.96
Prices include shipping, taxes, and discounts when available
Generated by PCPartPicker 2016-10-14 19:44 EDT-0400

AMD:
PCPartPicker part list / Price breakdown by merchant

CPU: AMD FX-8350 4.0GHz 8-Core Processor ($179.99 @ Newegg)
CPU Cooler: Cooler Master Hyper 212 EVO 82.9 CFM Sleeve Bearing CPU Cooler ($29.99 @ Newegg)
Motherboard: MSI 970 GAMING ATX AM3+ Motherboard ($84.98 @ NCIX US)
Memory: G.Skill Ripjaws Series 8GB (2 x 4GB) DDR3-1600 Memory ($46.89 @ OutletPC)
Total: $341.85
Prices include shipping, taxes, and discounts when available
Generated by PCPartPicker 2016-10-14 19:44 EDT-0400

$43 more, slower, and requires a ton of tweaking to perform at that speed. And this is a cheap example, if you really want to keep up you'll need liquid cooling to get the best overclock which will cost you another $80 on top of the cooler I mentioned.
 
Solution

CTurbo

Pizza Monster
Moderator
In addition while you may hear that an FX-8350 can be heavily overclocked to be more competitive, it will cost you quite a bit more, between an expensive cooler, and a better quality motherboard you will spend nearly twice as much to get less performance than a stock i5-6500 on a $50 motherboard using the free cooler that comes in the box.


Amen to this^^^
 
RL covered some good points. In addition to what he says, realize that synthetic benchmarks are good for seeing potential of hardware, but not real-world performance. They're programmed to take advantage of every single resource available to CPUs and such. But real world software isn't. So even though an i7 will be vastly superior to an i5 in benchmarks, it's usually not a big difference when running typical games and such. Same will go when comparing an FX-8300 against a 6300.

Now, if you're looking to build a low-budget gamer, how much money can you spend and what country will you be buying in? Answer these questions and we can help point you in the right direction for what parts to get.