Using average of those two vs aggregate score linked below, +21% ST, +22% MT over 7800x3d5.25 1t
BIOSTAR Group X870E VALKYRIE - Geekbench
Benchmark results for a BIOSTAR Group X870E VALKYRIE with an AMD Ryzen 7 9800X3D processor.browser.geekbench.com
5.3 1t
ASUS System Product Name - Geekbench
Benchmark results for an ASUS System Product Name with an AMD Ryzen 7 9800X3D processor.browser.geekbench.com
Make of that what you will.
Personally, I don't think GB6 is the correct benchmark from which to extrapolate X3D performance.Using average of those two vs aggregate score linked below, +21% ST, +22% MT over 7800x3d
So don’t extrapolate it? The numbers are what they are. You need to layer on your own temperaments, considerations, logic. Doesn’t make GB6 output invalid. It’s the interpretation and application that matters.Personally, I don't think GB6 is the correct benchmark from which to extrapolate X3D performance.
Zen 5 in general showed improvements there but not gaming. And X3D are for gaming.
Yes, exactly. I am cautioning against such extrapolation. Don't expect 9800X3D to be that much better in what it is actually built to do.So don’t extrapolate it? The numbers are what they are. You need to layer on your own temperaments, considerations, logic. Doesn’t make GB6 output invalid. It’s the interpretation and application that matters.
God forbid id want my single ccd gaming focused chip to excel in any way at (de)compression, program installation, transfer read/write….Yes, exactly. I am cautioning against such extrapolation. Don't expect 9800X3D to be that much better in what it is actually built to do.
eh it's not your fault, he really should do better than flip-flopping between compilers.Regarding SPEC, I will stop posting Geekerwan till they fix this.
With DDR5-4800 memory? Who would do that?9800x3d overall crossmark score 8.6% higher than 7800x3d
View attachment 110211View attachment 110212
It's a JEDEC employee.With DDR5-4800 memory? Who would do that?
It's a JEDEC employee.
eh it's not your fault, he really should do better than flip-flopping between compilers.
I'll bet he has multiple people doing benchmarks including running SPEC for him for his reviews, and they are all making their own choices. It doesn't make sense that he'd be running it himself and switching back and forth. He needs to get his helpers on the same page.
Zen 4 and 5 support the DDR5-5600 JEDEC standard though. . .
DDR5-5600 would be understandable, but 4800 - is really not. You can't even save any measurable amount of money between them.
BTW, good news with 9800x3d is official, out of the box support for DDR5-6000, which also shows the CPU in good light.
Probably quite conservative 5.2 GHz clock speed, (combined with official support for DDR-6000) IMO make 9800x3d a good candidate for the Gaming PC OEMs (such as Alienware) to finally switch to AMD.
Alienware? As in owned by Dell Alienware? I doubt that.
Comparison: https://browser.geekbench.com/v6/cpu/compare/8429253?baseline=8464281Using average of those two vs aggregate score linked below, +21% ST, +22% MT over 7800x3d
And this is vs a ‘tuned’ 7800x3d with decent ram, +13% ST and +4% MT, but without knowing how OCd the zen 5 was hard to know if it’s an unfair comparison.
Big deal let’s see it beat threadripper9800X3D is brave enough to even take on an average 7950X3D!
lol it is impressive... TR is not in the same price range.Big deal let’s see it beat threadripper
If you have a spare SSD, try Win11 24H2 just for benchmarking. I think you will easily breach 40k in Dawntrail.Gained 30% on the Dawntrail bench, didnt even re-install Windows, it just booted right back up with zero issues so I just rolled with it, lol.
Win10 22H2 is actually fastest in dawntrail, both for AMD and Raptor Lake 🧐 (~500points difference at the highend)If you have a spare SSD, try Win11 24H2 just for benchmarking. I think you will easily breach 40k in Dawntrail.
Still wondering what this setting actually does.. I can see higher gaming performance with disabling CCD1/SMT, but higher memory bandwidth ?!I tried out GIGABYTE X3D Turbo Mode very briefly and tested it with Sandra and Forza (9700X, D5-6000 16Gx2) ・SMT was turned off ・Memory bandwidth also became slightly thicker (about 55GB/s → about 61GB/s) ・No change in PPT (88W with CB R23) ・CPU simulation and rendering FPS have increased significantly ・As a result, the average frame rate increased by about 5% (Will it increase even more with X3D?)