• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

APU results Just Cause 3

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Enigmoid said:
http--www.gamegpu.ru-images-stories-Test_GPU-APU-Just_Cause_3-2.jpg


I hate how their bar lengths seem arbitrary. Why is 22 fps on the 4770K given a longer bar than 27 fps on the 7870k?
Click to expand...
Someone must put the Pentium and the i3 Haswell and Skylake to confirm that... Is really badly made for AMD APU's... To make it worse, seems that they (Iris Pro) are ALREADY BETTER compared to some low tier dGPU up to GTX 750 or R7 260.

BTW... Found this

http--www.gamegpu.ru-images-stories-Test_GPU-APU-Just_Cause_3-1.jpg


I fell that Celerons, Pentium and even i3 and i5 U tier and i3 HW desktop chips will have a really awful time in this game.
 
Last edited:
dark zero said:
I fell that Celerons, Pentium and even i3 and i5 U tier and i3 HW desktop chips will have a really awful time in this game.
Click to expand...
You have to realize that core load is not actual threads running,it's core load OVER TIME,task manager juggles all available threads around on all available cores so as to not overheat/wear down the first core.
 
monstercameron said:
wouldnt be surprised if he has some kind of relationship with the mods here. Alot of other people have had many infractions yet it doesnt get any.
Click to expand...
ahahah, I love how you refer to the poster as IT! ahahha :thumbsup:

Infraction issued for thread crapping.
-- stahlhart
 
Last edited by a moderator:
2is said:
Something is clearly quite weak. What is it in your opinion?
Click to expand...

its been covered lots,

to small L1D,
to slow L2
poor associative L1i cache
The whole WCC and cache policy
the Really bad misspredict penalty
Not enough Integer/branch execution resources
the 5 cycle FPU latency thax to all FP units having FMA

thats a pretty good starting point.
the funny thing is to me it seems like the architectural goals and direction was just wrong and that masks all the good work that AMD did going form STARS to CON.
 
The memory config is simply listed as "8Gb" is that 2x4 or 1x8, because that makes a big difference. And with that strange chart stacking min+avg, that's a new one.
 
xorbe said:
The memory config is simply listed as "8Gb" is that 2x4 or 1x8, because that makes a big difference. And with that strange chart stacking min+avg, that's a new one.
Click to expand...

Dual channel and single channel do not have any big differences in performance for the vast majority of users, if that's what you meant.

You generally need synthetics/benches to tell any difference.
 
LTC8K6 said:
Dual channel and single channel do not have any big differences in performance for the vast majority of users, if that's what you meant.
Click to expand...

If we are talking about IGP performance, you're definitely wrong - dual channel will deliver a significant boost in such a case.
 
Last edited:
LTC8K6 said:
Dual channel and single channel do not have any big differences in performance for the vast majority of users, if that's what you meant.

You generally need synthetics/benches to tell any difference.
Click to expand...

What? APU integrated graphics framerates basically double with dual channel mem. That's the crazy part, so many single stick benchmarks on the web, almost like it's intentional.
 
monstercameron said:
wouldnt be surprised if he has some kind of relationship with the mods here. Alot of other people have had many infractions yet it doesnt get any.
Click to expand...

I actually would really like an option to add this guy to an ignore list. I'm surprised he gets so little infractions considering he is the biggest anti AMD hater here and throws out negativity every chance he gets.

Infraction issued for member callout.
--stahlhart
 
Last edited by a moderator:
behrouz said:
why do they bench at Res 1280x720?

Here dragon Dogma arisen at 1920x1080 With low setting:

http--www.gamegpu.ru-images-stories-Test_GPU-APU-Dragons_Dogma_Dark_Arisen-1.jpg
Click to expand...

Because dragon dogma's graphics are simplistic if not archaic so you can get ~30fps avg at 1080.
Look at the just cause results and imagine half the FPS,720p is the only sensible resolution at which you could play this game on a igpu.
 
JC3 is not representative of the general performance of a modern AMD APU.

It only takes comparing other modern graphic intensive games (Witcher 3, GTAV, newer Assassin Creed, Battle-Field 4 etc.) to see that this is an anomaly, and likely caused by a driver or bug issue.

I'm pretty sure if the JC3 developers worked with AMD on the performance issues it would change pretty dramatically.
 
This game runs denuvo btw. That possibly contributes to CPU issues.

2is said:
Increasing the demand on anything will not yield in better performance. If it's a CPU bottleneck and you increase settings, you are not reducing CPU load, you're merely putting an additional load on the GPU the CPU load will be there all the same. The best you can hope to do if you have GPU power to spare, is achieve better visuals without losing any additional performance, but you will not gain performance.

EDIT: Also, as it relates to integrated graphics, increasing any settings, be it a GPU loading or CPU loading one, will most certainly reduce performance because both the CPU and GPU have a shared resource, that being memory bandwidth. edram equipped IGP's go a long way to alleviate this, so it would likely be even more advantageous for Intel over AMD.
Click to expand...

well yeah thats what I was talking about. improved visuals at the same performance. For some people that would be worth it. especially since you are not putting 720p screens with a $400 CPU like the broadwell typically.

would have to be seen if the RAM can't cope.
 
kawi6rr said:
Thanks so much, this will make the forums enjoyable again.
Click to expand...

That's odd that you asked about the ignore feature, actually got a response about it, that you ACTUALLY wanted and didn't know about, and that was not a "good post". Someone may want to get that fixed as it seems there as a miscommunication. Maybe let Stalhart know that you actually didn't know about the ignore feature and needed that post.

I really don't see what more conversation about this topic will bring though. Or even that many people care. This just seems like a way to push your brand bias.
 
Has anyone tried this with an 860k and a low-end graphics card like a 240? If it's really the CPU at fault, we should see similar problems.
 
frozentundra123456 said:
Interesting results in this game. game.gpu results

Iris pro destroys fastest AMD apu, while even HD530 is faster, especially minumums. Have a feeling the AMD apus are being held back by the cpu.

Of course, overall it just shows how inadequate APUs really are for any kind of serious gaming. Even at 720p only iris pro is able to average 30FPS.
Click to expand...


For whatever it's worth, I have at least one game that as far as I can tell only uses a single core on my FX, and am still GPU bound at 19x12 on my 7970 @1050 / 1450MHz. I'd be surprised if two modules wasn't enough for an 866MHz 512SP, somewhat memory starved iGPU.

If not, AMD would probably have been better off lowering the iGPU clocks a bit to save on TDP.
 
xorbe said:
What? APU integrated graphics framerates basically double with dual channel mem. That's the crazy part, so many single stick benchmarks on the web, almost like it's intentional.
Click to expand...

Are that many people gaming with an IGP?

Interesting. Are there any benchmarks to show this framerate doubling merely from going to dual channel mode on the mobo?
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top