Razer Leviathan V2 gaming soundbar review
Guru3D NVMe Thermal Test - the heatsink vs. performance
EnGenius ECW220S 2x2 Cloud Access Point review
Alphacool Eisbaer Aurora HPE 360 LCS cooler review
Noctua NH-D12L CPU Cooler Review
Silicon Power XPOWER XS70 1TB NVMe SSD Review
Hyte Y60 chassis review
ASUS ROG Thor 1000W Platinum II (1000W PSU) review
ASUS ROG Rapture GT-AXE11000 WIFI6E router review
Backforce One Plus Gaming Chair review
AMD FX 8350 processor review
Posted by: on: 10/23/2012 03:56 AM [
83 comment(s) ]
We review the AMD FX 8350 processor today. You guys have been hearing the magic word 'Vishera' for a while now and it is the codename for the new Piledriver core based FX series processors from AMD. The CPU tested today has eight physical CPU cores, it will have a base clock of 4000 MHz, and can Turbo towards 4200. And that makes the FX series a multi-threaded high frequency monster. You can read our review right here at Guru3D of course.
Read article
Advertisement
pages « 2 3 4 5 > »
Kohlendioxidus
Senior Member
Posts: 1398
Senior Member
Posts: 1398
Posted on: 10/23/2012 10:23 AM
agree here. They should use standard res. To really test CPU power you have another set of benchies to do that.
Other review sites go much further as they actually search for single core Intel optimised 5 to 10 years old games just to show how bad a AMD cpu is. It just makes me
.
I guess a standard resolution (such as 1920x1080, 16AF, some antialiasing) would be beneficial for a comparison though as it tests other limitations of the platform (such as the integrated PCI-E controller vs chip based etc).
agree here. They should use standard res. To really test CPU power you have another set of benchies to do that.
Other review sites go much further as they actually search for single core Intel optimised 5 to 10 years old games just to show how bad a AMD cpu is. It just makes me
.
BlackZero
Senior Member
Posts: 8878
Senior Member
Posts: 8878
Posted on: 10/23/2012 10:28 AM
Performance seems to be good but power usage is still a little high, nothing unexpected though. A good step forward overall.
That would be 318watts on the CPU rails, which is a lot of power! but going by the almost 100w increase at a 500Mhz overclock it could prove significant. Though, most boards should allow you to disable OCP, but at those power levels (and heat) it's going to be tough on the circuitry so I would not expect many of these going over 5GHz anyway.
Overclocking aside, I would be more interested in seeing if there has been any improvements with frame latency times for 3D.
Performance seems to be good but power usage is still a little high, nothing unexpected though. A good step forward overall.
If, during this overclock, your system shuts down, that's a new protection from AMD. You've reached either the heat, TDP or OCP limit (this seems to happen once the CPU pulls more than 26.5A from the 12V CPU rails (8pin and 4pin connectors). If that happens decrease the voltage a bit.
That would be 318watts on the CPU rails, which is a lot of power! but going by the almost 100w increase at a 500Mhz overclock it could prove significant. Though, most boards should allow you to disable OCP, but at those power levels (and heat) it's going to be tough on the circuitry so I would not expect many of these going over 5GHz anyway.
Overclocking aside, I would be more interested in seeing if there has been any improvements with frame latency times for 3D.
Hilbert Hagedoorn
Don Vito Corleone
Posts: 44027
Don Vito Corleone
Posts: 44027
Posted on: 10/23/2012 10:37 AM
1) I really cannot understand WHY the heck they still test CPU's in games at 1024 resolution with lowest details possible? THIS IS STUPID
and
Perhaps open your eyes and actually look at the tests ? You'll notice results for 1024x768 - 1280x1024 - 1600x1200 and 1920x1200
Lower resolution scores however are included to show you the effect of CPU performance, as higher resolutions are GPU bound.
Lowest details possible ? FC2 is in DX10 mode with high quality settings and 4xAA - Crysis 2 uses DX11 with the HR texturepack in Extreme quality settings at 4xAA
http://www.guru3d.com/articles_pages/amd_fx_8350_processor_review,18.html
1) I really cannot understand WHY the heck they still test CPU's in games at 1024 resolution with lowest details possible? THIS IS STUPID
and
Perhaps open your eyes and actually look at the tests ? You'll notice results for 1024x768 - 1280x1024 - 1600x1200 and 1920x1200
Lower resolution scores however are included to show you the effect of CPU performance, as higher resolutions are GPU bound.
Lowest details possible ? FC2 is in DX10 mode with high quality settings and 4xAA - Crysis 2 uses DX11 with the HR texturepack in Extreme quality settings at 4xAA
http://www.guru3d.com/articles_pages/amd_fx_8350_processor_review,18.html
Texter
Senior Member
Posts: 3175
Senior Member
Posts: 3175
Posted on: 10/23/2012 10:39 AM
A single Tenerife or GK110 will shed some more light on Vishera performance soon enough.
A single Tenerife or GK110 will shed some more light on Vishera performance soon enough.
pages « 2 3 4 5 > »
Click here to post a comment for this article on the message forum.
Senior Member
Posts: 6643
I guess a standard resolution (such as 1920x1080, 16AF, some antialiasing) would be beneficial for a comparison though as it tests other limitations of the platform (such as the integrated PCI-E controller vs chip based etc).