Jump to content

FX 8350 better with SLI @4k than 4930k.

As someone who is using an FX-8350 on a 990 FXA-UD3, I don't think that you can call this comparison a champ.

In a price/performance ratio, yes, AMD is a strong contender, they always have been... but if you were to re-analyze with a 4770k/4790k, you'd likely have a much different result, especially if you overclocked both of them.

 

You can skew the results of a test to make anything a winner, i'm sure there's some way to compare a GeForce4 series card against a GTX 900 series card and have the GeForce 4 come out on top, it's all relative to what and how you're testing.

~Remember to quote posts to continue support on your thread~
-Don't be this kind of person-

CPU:  AMD Ryzen 7 5800x | RAM: 2x16GB Crucial Ripjaws Z | Cooling: XSPC/EK/Bitspower loop | MOBO: Gigabyte x570 Aorus Master | PSU: Seasonic Prime 750 Titanium  

SSD: 250GB Samsung 980 PRO (OS) | 1TB Crucial MX500| 2TB Crucial P2 | Case: Phanteks Evolv X | GPU: EVGA GTX 1080 Ti FTW3 (with EK Block) | HDD: 1x Seagate Barracuda 2TB

Link to comment
Share on other sites

Link to post
Share on other sites

So, conclusion - more ghz matter at higher res? They should do more testing with overclocking.

 

-snip-

But how would you explain the benchmark results? The only thing that AMD is better at is ghz and no hyperthreading to  possibly skew results.

Link to comment
Share on other sites

Link to post
Share on other sites

So, conclusion - more ghz matter at higher res? They should do more testing with overclocking.

 

But how would you explain the benchmark results? The only thing that AMD is better at is ghz and no hyperthreading to  possibly skew results.

How would you explain the results? uhh.. much the same way that was done here. raw result comparison.

~Remember to quote posts to continue support on your thread~
-Don't be this kind of person-

CPU:  AMD Ryzen 7 5800x | RAM: 2x16GB Crucial Ripjaws Z | Cooling: XSPC/EK/Bitspower loop | MOBO: Gigabyte x570 Aorus Master | PSU: Seasonic Prime 750 Titanium  

SSD: 250GB Samsung 980 PRO (OS) | 1TB Crucial MX500| 2TB Crucial P2 | Case: Phanteks Evolv X | GPU: EVGA GTX 1080 Ti FTW3 (with EK Block) | HDD: 1x Seagate Barracuda 2TB

Link to comment
Share on other sites

Link to post
Share on other sites

Repost from the news section. And spoiler alert. GPU bound scenario. Throw a celeron in there and it's going to perform the same within margin of error.

"I genuinely dislike the promulgation of false information, especially to people who are asking for help selecting new parts."

Link to comment
Share on other sites

Link to post
Share on other sites

So, conclusion - more ghz matter at higher res? They should do more testing with overclocking.

 

But how would you explain the benchmark results? The only thing that AMD is better at is ghz and no hyperthreading to  possibly skew results.

 

did you hear that? /whoosh....

 

Basically at 4k the CPU almost becomes redundant. It then comes down to the chipset and how it addresses and uses SLI.

 

And, from those results it's quite obvious that 990FX handles SLI better than X79 does. This doesn't come as a surprise, SLI on 990FX is very good -

 

http://www.tomshardware.co.uk/crossfire-sli-scaling-bottleneck,review-32668.html

 

So good that it outshines AMD's own multi GPU technology lol.

Area 51 2014. Intel 5820k@ 4.4ghz. MSI X99.16gb Quad channel ram. AMD Fury X.Asus RAIDR.OCZ ARC 480gb SSD. Velociraptor 600gb. 2tb WD.

Link to comment
Share on other sites

Link to post
Share on other sites

Repost from the news section. And spoiler alert. GPU bound scenario. Throw a celeron in there and it's going to perform the same within margin of error.

 

Tried it have you, oh super intelligent clever one?

Area 51 2014. Intel 5820k@ 4.4ghz. MSI X99.16gb Quad channel ram. AMD Fury X.Asus RAIDR.OCZ ARC 480gb SSD. Velociraptor 600gb. 2tb WD.

Link to comment
Share on other sites

Link to post
Share on other sites

 

And?

 

When news breaks it spreads. People light up their stupid social networking sites with the same thing, over and over.

 

Shit happens, man.

Area 51 2014. Intel 5820k@ 4.4ghz. MSI X99.16gb Quad channel ram. AMD Fury X.Asus RAIDR.OCZ ARC 480gb SSD. Velociraptor 600gb. 2tb WD.

Link to comment
Share on other sites

Link to post
Share on other sites

Tried it have you, oh super intelligent clever one?

I don't have to try it because I use logic and reasoning to come to the conclusion that it's a GPU bound scenario and the processor is irrelevant. Go read the original link in the news section. This has already been debated.

Added to ignore. Your close minded and unintelligible presence on this forum brings others down and spreads misinformation when others are trying to do good and help those who ask for it.

"I genuinely dislike the promulgation of false information, especially to people who are asking for help selecting new parts."

Link to comment
Share on other sites

Link to post
Share on other sites

Alright somebody wanna copy and paste my explanation of this here. I would but I'm currently on mobile.

Why do you always die right after I fix you?

 

Link to comment
Share on other sites

Link to post
Share on other sites

Tried it have you, oh super intelligent clever one?

 

Please lose the condescending attitude...

 

Did you even read that Tom's article? All it demonstrated is that SLI scaled better on AMD chipsets than Crossfire did. They both scaled better on the Intel solution (which wasn't X79 anyway).

 

It's as likely that this is due to less driver CPU overhead than anything else, and the test was (by the author's own admission) not run correctly anyway since he was testing multi-GPU configurations with FRAPS at a time when frame rate analysis was VERY important.

Link to comment
Share on other sites

Link to post
Share on other sites

When someone else breaks the CoC, the report button is located at the bottom of their post.

 

Oh I see. 

 

I don't report people, I simply ignore them and let them rant on. He ranted, when I refused to bite he insulted me. So, given he wants to play that game then hey, two can play etc.

 

Apologies Linus, I just have a different way of handling angry little people like him.

 

Any way, nice to see you in person as it were, hope life is treating you well.

Area 51 2014. Intel 5820k@ 4.4ghz. MSI X99.16gb Quad channel ram. AMD Fury X.Asus RAIDR.OCZ ARC 480gb SSD. Velociraptor 600gb. 2tb WD.

Link to comment
Share on other sites

Link to post
Share on other sites

Tried it have you, oh super intelligent clever one?

What I see here is simply a showcase that if you run your games at ultra your bottleneck is going to end up being the GPU not the CPU, whether your going to get results that always show the AMD on top will vary with mileage but as far as gaming goes your cpu is conspicuously overshadowed by GPU performance, I've run some of my own benchmarks and even at the lowest settings 640x480 in most of my "AAA" games I am still bottle necked by the GPU, so yeah I'm not surprised with this result, I would expect Intel to more consistently edge out AMD but I'm not ostensibly surprised.

 

What does surprise me is that Intel is consistently lower, I have seen up to 10fps variation in my own testing so I could see each test bouncing back and forth a frame or two but to have Intel consistently lower is a red flag to me that would suggest the result were true but were either cherry picked or simply randomly fell in AMDs favor. I have very little faith that the artificial benchmark results shown and feel that perhaps they should be run again.

Edit: As i've already said, we've played this exact song and dance once before and guess what @Faceman was there singing along to the same song he's singing right now, and because we as a group wrote the lyrics to this song a week ago we don't need to rewrite it with new benchmarks.

Why do you always die right after I fix you?

 

Link to comment
Share on other sites

Link to post
Share on other sites

What I see here is simply a showcase that if you run your games at ultra your bottleneck is going to end up being the GPU not the CPU, whether your going to get results that always show the AMD on top will vary with mileage but as far as gaming goes your cpu is conspicuously overshadowed by GPU performance, I've run some of my own benchmarks and even at the lowest settings 640x480 in most of my "AAA" games I am still bottle necked by the GPU, so yeah I'm not surprised with this result, I would expect Intel to more consistently edge out AMD but I'm not ostensibly surprised.

 

What does surprise me is that Intel is consistently lower, I have seen up to 10fps variation in my own testing so I could see each test bouncing back and forth a frame or two but to have Intel consistently lower is a red flag to me that would suggest the result were true but were either cherry picked or simply randomly fell in AMDs favor. I have very little faith that the artificial benchmark results shown and feel that perhaps they should be run again.

 

I actually have all of the tools and equipment to try it myself tbh. I've got an 8320 that does 4.9ghz and a 3970x that does the same. I got very lucky with both. I also have a pair of Titan Blacks, the only issue is the AMD rig has a 750w PSU and runs eight hard drives so it would be risky.

Area 51 2014. Intel 5820k@ 4.4ghz. MSI X99.16gb Quad channel ram. AMD Fury X.Asus RAIDR.OCZ ARC 480gb SSD. Velociraptor 600gb. 2tb WD.

Link to comment
Share on other sites

Link to post
Share on other sites

they should be run again.

This.

Even if it is gpu bound, intel should still win. As linus pointed out, benchmark methodology is problem and i would take results with salt mine of salt..

Link to comment
Share on other sites

Link to post
Share on other sites

I actually have all of the tools and equipment to try it myself tbh. I've got an 8320 that does 4.9ghz and a 3970x that does the same. I got very lucky with both. I also have a pair of Titan Blacks, the only issue is the AMD rig has a 750w PSU and runs eight hard drives so it would be risky.

You don't need too, this benchmark is mostly accurate, but if you would be willing to I would love to see the artificial benchmarks run again @4k so if you find time to do that I would be interested in seeing the results.

Why do you always die right after I fix you?

 

Link to comment
Share on other sites

Link to post
Share on other sites

At 4K you're going to be GPU bound. And these tests are a bit misleading also. Z97 is a much newer platform than 2011-2 and Z97 is the one most gamers are buying. Not a $500 4930K IVB-E and a X79 board.

 

I'd like these tests re-run on more accurate platforms with and without overclocks before they're plausible.

Link to comment
Share on other sites

Link to post
Share on other sites

This.

Even if it is gpu bound, intel should still win. As linus pointed out, benchmark methodology is problem and i would take results with salt mine of salt..

Not exactly, I've seen intel more consistently have higher min. framerates but the max ends up pretty close together and depending on how long you run the benchmark the averages will eventually start to coincide because the minimum frame rates are generally quick dips that become nothing more than a drop in the bucket. If the Gpu is pinned at 100% the CPU can't be let to stretch its legs, games are a race that can't be won by an individual, both parties have to cross the finish line at the same time so the CPU is forced to hold back and will make little to no difference in the end result as long as it partner is slowing him down, and vice versa, when the GPU is no longer the bottleneck the CPU can run as fast as it's allowed and that's when you will see differences between AMD and Intel (spoiler, Intel wins that race everytime)

Why do you always die right after I fix you?

 

Link to comment
Share on other sites

Link to post
Share on other sites

You don't need too, this benchmark is mostly accurate, but if you would be willing to I would love to see the artificial benchmarks run again @4k so if you find time to do that I would be interested in seeing the results.

 

Some one needs to.

 

I have a feeling that once you remove the CPU it's as simple as how well SLI scales and works. And it could just work better on 990FX.

Area 51 2014. Intel 5820k@ 4.4ghz. MSI X99.16gb Quad channel ram. AMD Fury X.Asus RAIDR.OCZ ARC 480gb SSD. Velociraptor 600gb. 2tb WD.

Link to comment
Share on other sites

Link to post
Share on other sites

The games that are tested in this review are all gpu bound games, in the first place.

But also at 4K you gonne be more gpu bound.

Link to comment
Share on other sites

Link to post
Share on other sites

Those benchmarks should use overclocked CPUs, I don't really see people buying more then one high-end graphic card and not overclocking their CPUs.

PC : | CPU: Intel 4790K | COOLER: Corsair H105 w/ JetFlo's Push/Pull | MOBO: EVGA Z97 Classified | GPU: EVGA FTW 4GB GTX 970 X2 | RAM: Kingston HyperX Beast 1866Mhz 32GB | CASE: HAF Stacker 945 | PSU: Corsair AX1500i | DISPLAY: Asus MX299Q | SSD: 2 X Corsair Neutron GTX 480 GB in RAID0 | mSATA SSD: Samsung 840 EVO 500 GB | HDD: 4 X Western Digital RED 4 TB in JBOD |

Link to comment
Share on other sites

Link to post
Share on other sites

What I see here is simply a showcase that if you run your games at ultra your bottleneck is going to end up being the GPU not the CPU, whether your going to get results that always show the AMD on top will vary with mileage but as far as gaming goes your cpu is conspicuously overshadowed by GPU performance, I've run some of my own benchmarks and even at the lowest settings 640x480 in most of my "AAA" games I am still bottle necked by the GPU, so yeah I'm not surprised with this result, I would expect Intel to more consistently edge out AMD but I'm not ostensibly surprised.

 

What does surprise me is that Intel is consistently lower, I have seen up to 10fps variation in my own testing so I could see each test bouncing back and forth a frame or two but to have Intel consistently lower is a red flag to me that would suggest the result were true but were either cherry picked or simply randomly fell in AMDs favor. I have very little faith that the artificial benchmark results shown and feel that perhaps they should be run again.

Edit: As i've already said, we've played this exact song and dance once before and guess what @Faceman was there singing along to the same song he's singing right now, and because we as a group wrote the lyrics to this song a week ago we don't need to rewrite it with new benchmarks.

Unigine heaven, he basically gained like 5 fps from 780 SLI to 980 SLI, where as he gained 25 fps with the 8350.

http://imagescdn.tweaktown.com/tweakipedia/5/5/55_53_amd_fx_8350_powering_gtx_780_sli_vs_gtx_980_sli_at_2560x1440.png

http://www.pcper.com/files/imagecache/article_max_width/review/2014-09-18/G3-heaven.png (3960x)

It's clear what kinda reviewer it is though but whatever. Let me give you an idea; http://be.hardware.info/reviews/5109/24/amd-fx-9590--fx-9370-review-amds-rentree-in-de-high-end-markt-benchmarks-hd-7970-tomb-raider-1920x1080-medium

When you're hitting a GPU limitation, weirdest thing can happen like in that hardware.info result. You can't use results like this to claim that x is better than y, what this reviewer has been doing.

Link to comment
Share on other sites

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×