GTA5 CPU Benchmarks (I3 beats FX yet again)

Martini1991 · 25 Apr 2015 at 14:42

willhub said:
How poor?

Any comparisons?

Probably somewhere between that 4GHZ Phenom OC and the FX4300 stock.

You're basically running 4 Yorkdale cores @ 3.7 aren't you?

gpuerrilla · 25 Apr 2015 at 14:48

OK a post that seems helpful found:

C35799Gamble said:
I have managed to stop tinkering with settings now and am getting a solid 60FPS, with good temps so I don't need to monitor it anymore.

Im currently using:
Res: 1920x1080
Vsync in game Off
Vsync in Nvidia control panel - adaptive

All settings on High
FXAA - On
MSAA - off
Reflection MSAA - off
all three draw sliders - mid way
Antistropic filtering x16
Post fx - Normal(only disables motion blur which is fine by me)

This gives me 60FPS all day long with no drops(unless im online then it drops to about 57fps) with temps sitting around 50 for CPU/GPU1 and GPU2.

System:
CPU: AMD fx8350 @ 4.4
RAM: 8gb
GFX: Nvidia 770 4GB x2
SSD: Samsung EVO

I tested everything on very high and still got 60FPS consistently but my temps on both cards were breaching 70 and considering I have been playing for hours on end, I don't want them sitting at those temps for that length of time so put the settings down to high instead.

This is based on:

* a real in game test by a persons system - not a benchmark
* the game being playable
* The FX actually not being overclocked to its potential (an 8350 should be capable of 4.7Ghz comfortably)
* note this is NOT an i3!

willhub · 25 Apr 2015 at 14:50

Martini1991 said:
Probably somewhere between that 4GHZ Phenom OC and the FX4300 stock.

You're basically running 4 Yorkdale cores @ 3.7 aren't you?

Yeah.

So pretty atrocious performance?

I'll have a HD7970 3GB in the rig.

humbug · 25 Apr 2015 at 14:51

Martini1991 said:
Biased.
Happy Spoffle ?

Most of his claims are BS (Said as much previously), but I still find others to be as bad/worse. While Dave can talk crap (And does), there's an absolute truth in Intel have the higher performing CPU's, and this can come through in games, and while he may try and exaggerate, there's plenty of AMD owners who simply ignore it and pretend there's no truth in anything he'd say. Given the choice these people would take an FX83 over a modern i5 in gaming, despite the i5 being factually better. In order for one to do so, a bias must exist.

Given the choice these people would take an FX83 over a modern i5 in gaming, despite the i5 being factually better. In order for one to do so, a bias must exist.

Infantile arrogance.

People chose an FX-83## because to them they are better, whether you agree or not.

Its because they don't agree with you that makes them in your mind bias.

Martini1991 · 25 Apr 2015 at 14:51

I wouldn't be expecting miracles.
But with a 7970 you're bottlenecked in nigh on every game you play anyway

willhub · 25 Apr 2015 at 14:52

Martini1991 said:
I wouldn't be expecting miracles.
But with a 7970 you're bottlenecked in nigh on every game you play anyway

7970 will be strangled by the CPU?

Oh dear, looks like CPU/MOBO/RAM upgrade, could probably budget 150 pound for that, not up to speed with processors and all that I'm 7 years behind

Martini1991 · 25 Apr 2015 at 14:53

willhub said:
7970 will be strangled by the CPU?

Oh dear, looks like CPU/MOBO/RAM upgrade, could probably budget 150 pound for that, not up to speed with processors and all that I'm 7 years behind

You're going to have to go second hand.

Martini1991 · 25 Apr 2015 at 14:55

humbug said:
Infantile arrogance.

People chose an FX-83## because to them they are better, whether you agree or not.

Its because they don't agree with you that makes them in your mind bias.

So, a 8 core Piledriver is better than a modern i5 in gaming overall, yes or no? There's no opinion, it's fact. And the i5 is the better CPU overall in games.
To choose one in spite of the facts is biased.

That's logic rather than arrogance.

humbug · 25 Apr 2015 at 14:59

Martini1991 said:
So, a 8 core Piledriver is better than a modern i5 in gaming overall, yes or no? There's no opinion, it's fact. And the i5 is the better CPU overall in games.
To choose one in spite of the facts is bias.

That's logic rather than arrogance.

It is for me yes.

gpuerrilla · 25 Apr 2015 at 14:59

humbug said:
Infantile arrogance.

People chose an FX-83## because to them they are better, whether you agree or not.

Its because they don't agree with you that makes them in your mind bias.

Good observation. The earlier post made against Andy was another example.

Of course if you post in a thread you are always defending AMD to the hilt and not obliged to reply if you think the information presented is inaccurate.

I enjoy a technical debate. It seems the same faces rarely bring to the table external facts to support themselves. It just comes across as amateurish bandwagon slurs when you get to the nitty gritty.

Martini1991 said:
So, a 8 core Piledriver is better than a modern i5 in gaming overall, yes or no? There's no opinion, it's fact. And the i5 is the better CPU overall in games.

This is the usual backed into a corner remark, no one contests this martini - however you use it like the old top trumps cards spoilt kids used in the playground. Funnily enough this thread wasn't about i5's but then again if you want to argue that old chestnut again you will find they belong in a different price bracket and generally have games optimised to run on their hardware to stand out!

Martini1991 · 25 Apr 2015 at 14:59

humbug said:
It is for me yes.

I don't know what to say now.

Martini1991 · 25 Apr 2015 at 15:14

Th0nt said:
OK a post that seems helpful found:

This is based on:

* a real in game test by a persons system - not a benchmark
* the game being playable
* The FX actually not being overclocked to its potential (an 8350 should be capable of 4.7Ghz comfortably)
* note this is NOT an i3!

How is this good?

Using those settings I managed the 60 FPS with a lower GPU set up (Single 290X) at a higher resolution (2560x1080)

I was curious, so tested it.

Th0nt said:
This is the usual backed into a corner remark, no one contests this martini - however you use it like the old top trumps cards spoilt kids used in the playground. Funnily enough this thread wasn't about i5's but then again if you want to argue that old chestnut again you will find they belong in a different price bracket and generally have games optimised to run on their hardware to stand out!

Well, actually no. Because Humbug just contested it. And again, wasn't coming from a backed into a corner remark (To say so would imply I'm arguing from a position of weakness, I disagree)

humbug · 25 Apr 2015 at 15:18

When you turn up the Quality settings in a game you're adding Physics, Lighting, Shading.... calc which is where low threaded performance becomes less important and high threaded performance more important.

Martini1991 · 25 Apr 2015 at 15:19

humbug said:
When you turn up the Quality settings in a game your adding Physics, Lighting, Shading.... calc which is where low threaded performance becomes less important and high threaded performance more important.

Assuming it's threaded well. If not, it's the opposite.

So, an almost entirely irrelevant statement.

gpuerrilla · 25 Apr 2015 at 15:23

Martini1991 said:
Assuming it's threaded well. If not, it's the opposite.

So, an almost entirely irrelevant statement.

So is GTA5 well threaded or not? I heard it was.

Martini1991 · 25 Apr 2015 at 15:24

Th0nt said:
So is GTA5 well threaded or not? I heard it was.

Irrelevant to the statement, given the statement said "A game".
It was just another blanket statement which helps no one.

It could very well be the truth with GTA 5, but I've not spent much time messing with it looking at CPU usage etc.

Tell you what, I'm bored, I'm doing nothing, stuck in a rut, keeping an eye on BBC's coverage of the Boro game. I'll sort my MSI AB out and run some tests for you

humbug · 25 Apr 2015 at 15:27

Martini1991 said:
Assuming it's threaded well. If not, it's the opposite.

So, an almost entirely irrelevant statement.

Few Engines these days run all calculations on just a few threads, actually i will rephrase that, no DX11 Engine run all calculations on just a few threads.

Its not the Game Engine which decides how and where it calculates what, its the API the engine is built around.

Having said that not all games require a higher degree of calc.

humbug · 25 Apr 2015 at 15:49

Tell ye' what.

In terms of performance i would have an FX-63## over any i3
For performance its a close call between a Haswell i5K and an FX-83##, For cost, Value for money its the FX-83## hands down.

For performance a Haswell i7 is a juicy prospect, but the price at x2 that of an FX-83## is bonkers, the i7 price IMO compare it to nothing is just plain bonkers.

But here is the cracker, what i really want is an Xeon 5650, more than a Haswell 4c 8t i7, i tried like hell to get one when they were going for £65 but a good X58 motherboard was hard to get and very expensive, the price of that CPU has now gone up, as have the boards. £150 for a good used one.

gpuerrilla · 25 Apr 2015 at 15:58

Martini1991 said:
Tell you what, I'm bored, I'm doing nothing, stuck in a rut, keeping an eye on BBC's coverage of the Boro game. I'll sort my MSI AB out and run some tests for you

Great, you also reminded me of the games going on so switching to the site tab myself!

ALXAndy · 25 Apr 2015 at 16:05

Martini1991 said:
So, a 8 core Piledriver is better than a modern i5 in gaming overall, yes or no? There's no opinion, it's fact. And the i5 is the better CPU overall in games.
To choose one in spite of the facts is biased.

That's logic rather than arrogance.

To be quite bloody honest, Martini, in any modern game there is nothing between any CPU in gaming. Literally 0-5 FPS between a FX 8320 and a 5960x.

Because let's face it threading and specific CPU support is time and time is money.

CPUS hardly even matter any more and there is no reason to over spend on one if gaming is your sole intention. If I had say £500 to spend I would buy a 6300 and dump the rest into the best GPU I could get my hands on.