• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

AMD Zen 2 (Ryzen 3000) - *** NO COMPETITOR HINTING ***

I will wait until they put it on their official website where I got the ABBA BIOS from. No rush, perfectly happy rocking my 4.4GHz all core OC with this BIOS :D
 
What are your settings for 4.4Ghz?
No settings. Default BIOS with only change being XMP memory.

I use Ryzen Master for the OC leave it at 1.275v. To me that seems to be the sweet spot as when I look at when the thew stock boosts when running Cinebench, it seems to want to use about that much. The only time it uses more vaults is on single core boost or lower loads.

Right now I am experimenting with different clocks and voltages and update my results now and then on the Ryzen 3600 thread. You can also take a look at the Cinebench Benchmark thread to see some of my results where I even manages 4.6GHz, but that was only on 3 cores.

Next I will be testing my 3 cores at 4.5GHz and 3 at 4GHz profile at 1.275v and see how that goes for a while. But I think I would rather a all core 4.4GHz as that gives more performance overall and only lose on 100MHz single core performance.
 
Is it me or is 1.4v for 3700x a little high?
Also what stress test you all using
I use Cinebench, Blender Benchmark, 3DMark, AIDIA64, OCCT and play games. If it does not crash then it's all good as far as I am concerned :D

Being honest all you need is Cinebench and playing games and using your PC as normal, that is what I mainly do. If it does not crash then it's all good.
 
I use Cinebench, Blender Benchmark, 3DMark, AIDIA64, OCCT and play games. If it does not crash then it's all good as far as I am concerned :D

Being honest all you need is Cinebench and playing games and using your PC as normal, that is what I mainly do. If it does not crash then it's all good.
I’m using cinebench but the problem with that is that it doesn’t go on for long enough.
With my 2500k I used intel burn test which game me enough time to set up my fans for when it runs hot and then fine tune as I use.

Cinebench gave loads of time with my 2500k with a score in the 900
 
I’m using cinebench but the problem with that is that it doesn’t go on for long enough.
With my 2500k I used intel burn test which game me enough time to set up my fans for when it runs hot and then fine tune as I use.

Cinebench gave loads of time with my 2500k with a score in the 900
Use Blender Benchmark then. That takes at least 20 minutes to run on the quick test. If you want longer then run the full test ;)

Also use OCCT and run it as much as you want or AIDA64.
 
Few benchmarks start appearing with AGESA 1004. The AC performance gains are truly impressive, 10% flat almost!!!

 
AC performance gains are truly impressive
If you play at a potato resolution. At 4k it will make zero difference :p

I just hope my current OC does not lose stability when I update as it would be nice to have faster booting and all the other small fixes.
 
Low resolution on a 2080TI is perfectly legitimate when testing CPU performance.

Higher resolution makes the GPU, not the CPU the bottleneck, if you want to know how good the CPU can handle games then having it limited by GPU performance is completely counter productive, in fact it becomes a GPU, not CPU review.
 
Tell that to the discussion here, where today is all about fps at 720p & 1080p med/high settings using RTX2080TI :rolleyes:
Only way to get Ray Tracing working at playable fps. Lol.


Low resolution on a 2080TI is perfectly legitimate when testing CPU performance.

Higher resolution makes the GPU, not the CPU the bottleneck, if you want to know how good the CPU can handle games then having it limited by GPU performance is completely counter productive, in fact it becomes a GPU, not CPU review.
I am sure he knew this humbug :D
 
Yeah that's a 10% jump in IPC, 1003 ABBA 4.225Ghz, 1004 4.2Ghz and clearly 10% higher FPS in AC, WTF??????
Almost certainly just the monitoring software not understanding the clock speeds. Remember Ryzen clock speeds can adjust extremely quickly and in 25 MHz increments. The performance difference is ridiculous if this is accurate but I am not convinced.
 
Low resolution on a 2080TI is perfectly legitimate when testing CPU performance.

Higher resolution makes the GPU, not the CPU the bottleneck, if you want to know how good the CPU can handle games then having it limited by GPU performance is completely counter productive, in fact it becomes a GPU, not CPU review.

Playing at lower resolution with a 2000 dollar card to see Wich cpu pulls a higher fps when in reality you will use 1440p ultra its like saying let's play games and streams simultaneously to see wich cpu pulls out higher fps.
 
Almost certainly just the monitoring software not understanding the clock speeds. Remember Ryzen clock speeds can adjust extremely quickly and in 25 MHz increments. The performance difference is ridiculous if this is accurate but I am not convinced.

Didn't you both notice that despite the lower clocks, the Agesa 1.0.0.4 keeps all the cores loaded. So, it isn't about the frequency but about how the game threads are distributed and executed by the CPU.
 
Didn't you both notice that despite the lower clocks, the Agesa 1.0.0.4 keeps all the cores loaded. So, it isn't about the frequency but about how the game threads are distributed and executed by the CPU.
Yep. Might seem weird what I will say, but feeling that with AGESA 1003ABBA & 1004 AMD is doing half the job of what the windows scheduler should be doing and half the job of developers.
At least with AMD GPU drivers where AMD can properly program to work in sync.

Why I wrote that. X4 Foundations, on my 3900X + 5700XT AE uses 10 cores at 60-80% (2.6 patch). Others reported on the forums using 4 cores on 3900X + GTX1070 and all of them at almost 100%.
However haven't got a confirmation which patch the guy was using.
 
Back
Top Bottom