[Tom's] CPU Bottlenecking with 7970 CF - 3770k vs. 8350

Page 6 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

AtenRa

Lifer
Feb 2, 2009
14,003
3,362
136
And I have proven to you with fraps recordings and screenshots that the ingame benchmark does in fact not represent accurate data as far as real gameplay is concerned. I have pointed out that Shogun 2 doesn't record camera movements and have expressed doubt that you could have manually moved the camera in the exact same manner for 300+ seconds. Thus your benchmark data is bogus.

And yet, you keep ignoring it, just further proving your ignorance and learning-resistance when it comes to benchmarking CPUs.

It is simple, do the same pattern every time, try to choreograph the way you move the camera in 5-6 (or more) movements. Try to do the same movements every time you run the bench.

ok listen, do your own reviews the way you feel its right. I will do mine the way I believe its right.

You keep calling me just because you disagree, as you have seen I havent call you anything. Now, go bench your games, make the graphs and upload your own review and stop calling people when you have different opinions.
 

boxleitnerb

Platinum Member
Nov 1, 2011
2,605
6
81
It is simple, do the same pattern every time, try to choreograph the way you move the camera in 5-6 (or more) movements. Try to do the same movements every time you run the bench.

ok listen, do your own reviews the way you feel its right. I will do mine the way I believe its right.

You keep calling me just because you disagree, as you have seen I havent call you anything. Now, go bench your games, make the graphs and upload your own review and stop calling people when you have different opinions.

I will call you on flawed data as I see fit, thank you. In a real battle, fps will be far lower than what you show, unless you're looking away from the action.
Why I know that? Because I've actually played the game for months. I doubt you have, otherwise you wouldn't make such ridiculous claims.
 

AtenRa

Lifer
Feb 2, 2009
14,003
3,362
136
. In a real battle, fps will be far lower than what you show, unless you're looking away from the action.

The problem with you is that you actually believe that every player will play the same game like you do. Well bad news for you, every player will play the same game different than the others.
You may play with more zoom in, others may play at mind zoom etc. The in game benchmark runs scenes both in zoomed in and zoomed out, so you have an average of the performance you will get playing the game.

Anyway, lets end it here.
 

boxleitnerb

Platinum Member
Nov 1, 2011
2,605
6
81
The ingame benchmark displays only a small number of units at any given time compared to a real battle. Also, there are no engagements whatsoever, so no combat calculations, no collisions and most importantly (since it is pre-recorded) no AI that is reacting to player input. So no, it is not the same.

And of course every player plays in a different way. However, if conducting CPU benchmarks that others may include in their purchasing decision, it is better to use more demanding ways/scenarios (that are still representative) so that there won't be nasty surprises later. When benchmarking graphics cards, you use graphically intensive games/settings to judge their performance. The same should apply when it comes to CPUs.
 
Last edited:

dawp

Lifer
Jul 2, 2005
11,347
2,710
136
AMD motherboards are also cheaper than Intel's so a $150 savings is doable.

This is the cheapest ATX motherboard that has a USB3 header.

PCPartPicker part list / Price breakdown by merchant / Benchmarks

CPU: AMD FX-6300 3.5GHz 6-Core Processor ($129.99 @ Microcenter)
Motherboard: Asus M5A97 R2.0 ATX AM3+ Motherboard ($84.99 @ Amazon)
Total: $214.98
(Prices include shipping, taxes, and discounts when available.)
(Generated by PCPartPicker 2013-01-24 23:34 EST-0500)

if you're going to go through the trouble of going into a Microcenter store to get the cpu since that $129 price is in store only, you might as well get a bundle and save $40+ depending on what board you chose.
http://microcenter.com/site/products/amd_bundles.aspx
 

Puppies04

Diamond Member
Apr 25, 2011
5,909
17
76
As of today, 1 GB Pound = 1.58 USD

£35 = ~55.00 USD with that you can get a 80+ 450W PSU like the COUGAR RS450.

£40 = ~63.00 USD with that you can get a 970 chipset AM3+ motherboard like the ASRock 970 PRO3 AM3+

No need to add anything ;)

If you are trying to convert £ into $ at a flat exchange rate and assuming that hardware costs the same in the USA and the UK you are extremely misguided.

Also considering I was specifically talking about OCing an FX 6 core (a CPU family known for using a lot of power when OC'd) and we are assuming usage of a 7870 I certainly would want more than a 450W PSU for long term use.
 
Last edited:

AtenRa

Lifer
Feb 2, 2009
14,003
3,362
136
If you are trying to convert £ into $ at a flat exchange rate and assuming that hardware costs the same in the USA and the UK you are extremely misguided.

Gigabyte 970A-DS3 = £59.99 inc VAT

Corsair Builder Series CX 430W V2 '80 Plus Bronze' = £36.98 inc VAT

I would strongly suggest you spend more than £40 on a mobo and £35 on a PSU of course once you add £20 to each of those items

So, we are coming to the same conclusion, you dont need to add anything more either you buy in the US or the UK ;)
 

AtenRa

Lifer
Feb 2, 2009
14,003
3,362
136
Also considering I was specifically talking about OCing an FX 6 core (a CPU family known for using a lot of power when OC'd) and we are assuming usage of a 7870 I certainly would want more than a 450W PSU for long term use.

You can OC the FX6300 to 4GHz without needing more than the 450W PSU (with the HD7850).
 

Puppies04

Diamond Member
Apr 25, 2011
5,909
17
76
You can OC the FX6300 to 4GHz without needing more than the 450W PSU (with the HD7850).

I believe I was originally quoting a message with regards to a 7870 not 7850 and while I would rather have a FX6300 than an I3 (at the same complete system price) that would only apply if the FX was OC'd to at least 4.5 at which point I would certainly expect to need more than a 450W PSU.
 

Piroko

Senior member
Jan 10, 2013
905
79
91
A quality 450W psu should still be plenty. The 7870 is 175w tdp, together with the rest of the system that would still leave 200W for the cpu. And the Piledriver power consumption isn't that horrible even at 4.5 Ghz.
 

AtenRa

Lifer
Feb 2, 2009
14,003
3,362
136
I believe I was originally quoting a message with regards to a 7870 not 7850 and while I would rather have a FX6300 than an I3 (at the same complete system price) that would only apply if the FX was OC'd to at least 4.5 at which point I would certainly expect to need more than a 450W PSU.

Your quote,

Replace that 6 core AMD with an I3 and you have exactly the same kind of value. Sure you can OC the AMD rig but if you are planning on doing that I would strongly suggest you spend more than £40 on a mobo and £35 on a PSU of course once you add £20 to each of those items then you are within about £30 of being able to afford an unlocked I5 which will quite frankly blow the I3 or 6 core FX out of the water even at stock.

FX6300 at 4GHz will smoke the Core i3, you dont need to OC at 4.5GHz. But even if you like to OC to 4.5GHz including the HD7870, the Corsair CX 430W PSU will be sufficient.
 

Piroko

Senior member
Jan 10, 2013
905
79
91
The improved FP unit from Piledriver would help in some games, but overall I would consider it a sidegrade, not worth doing.
 

guskline

Diamond Member
Apr 17, 2006
5,338
476
126
I too have a 7870 and a 8150, would a switch to 6300 benefit any?
Hard to answer your question as I never had a 6000 series cpu. The 8350 is an improvement over the 8150. However, I'm inclined to agree with the poster who said at most a side grade.