Post your Cinebench R23 Score

[FCG]

It is my personal PC, but on occasion I bring work home: Isotopes, nuc med, medical images, deep learning algorithms (ANN CNN), spreadsheets, email, and gaming.

The chips are QS.

Ohhh... what do you think you know about isotopes. We should take this offline.

 

[Apocalypsee]

Ryzen 5 1600 3.7GHz (locked it at boost clock in BIOS) with Arctic Freezer 64 Pro (yes cooler from Athlon 64 days) with 2x8GB Samsung generic B die 2133MHz @ 3133MHz.

 

[sam_86314]

Here's a system I've been working on that I haven't posted about yet (will write about it soon).

i3 4130 (stock speed), stock Intel cooler

 

[natr0n]

air cooling

 

[Melvis]

Intel i5-6300HQ @ Stock (2.8GHz) , 8GB DDR4 2133MHz, Air Cooled (Lenovo Y700)

 

[Enterprise24]

8700K @ 5.2Ghz delidded + custom loop + 31C ambient.

 

[sam_86314]

Used an SSD, 8GB of memory, and Windows 10 to rejuvenate my old laptop from my first year of high school; an HP G62-347CL.

Athlon II P340 (stock speed), air cooled.

 

[steen]

BTW, your system, once normalized for core count to my system, means the dual Xeon Scalable 8180(M) provides for only a 3.3% increase in compute power (in this limited instance) over the circa-2015 dual Haswell-EP E5-2696 v3 when properly *ahem* unlocked.

UEFI turbo microcode patch? 2696V3 @ the right price is a relative bargain.

To OP & members: Can we also post single thread results for comparisons, esp from the high core count/interesting cpu crowd.

 

[ArbitraryAffection]

Used an SSD, 8GB of memory, and Windows 10 to rejuvenate my old laptop from my first year of high school; an HP G62-347CL.

Athlon II P340 (stock speed), air cooled.

Wow what a Beast.

To OP & members: Can we also post single thread results for comparisons, esp from the high core count/interesting cpu crowd.

I will run it right away for you sir!

I did some tweaking to my PBO settings in Ryzen master and i can squeeze an extrra 50-100 mhz from the 2700x at stock with no manual oc even on my noctua 120mm air cooler. I will multi again to see if it made a huge difference~

update:

I ran Both tests with my 2700X and new tweaked PBO settings. Everything else is the same:

I am super happy with this result:3

I also ran it second time on both tests to see what the average clock rates were, according to HWINFO64. So in the Multi-core test, all the 8 cores were at ~4041 MHz.

In the single cores test, the core being used was around ~4290 MHz.

It was at 70*c and my fans on CPU cooler are not spinning at maximum load. Again this is not a manual oc but Ryzen master set to this:

 

[Vycyous]

Ran with everything, except memory, at stock settings (no tweaks, PBO overclocking, etc.). Specs listed on profile. Ambient temperature of about 20°C, case side panel on. Memory is running at 3333 MT/s 14-14-14-30 1T (as you can see in CPU-Z).

 

[tvamos]

Air cooler, R5 1600 @3816MHz, 16GB @3066MHz

 

[ArbitraryAffection]

Ran with everything, except memory, at stock settings (no tweaks, PBO overclocking, etc.). Specs listed on profile. Ambient temperature of about 20°C, case side panel on. Memory is running at 3333 MT/s 14-14-14-30 1T (as you can see in CPU-Z).

View attachment 118315

nice our scores are very similar. i tweaked my settings in RM to achieve my score, at stock it may be because my motherboard vrm is a lot weaker than yours. And i have a smalle cooler so i will hit the temp that drops my boost bin down a notch quicker.(I use Asrock b450M pro4 btw). But overal we have bang near the same score. your slight edge in memory bandwidth though. might i ask what your memory tRFC is please? Mine is 560T. my kit is in my specs, totally XMP stock. Just wondering if it worth maybe dropping my tRFC down. I heard it can improve it a bit.

 

[Vycyous]

nice our scores are very similar. i tweaked my settings in RM to achieve my score, at stock it may be because my motherboard vrm is a lot weaker than yours. And i have a smalle cooler so i will hit the temp that drops my boost bin down a notch quicker.(I use Asrock b450M pro4 btw). But overal we have bang near the same score. your slight edge in memory bandwidth though. might i ask what your memory tRFC is please? Mine is 560T. my kit is in my specs, totally XMP stock. Just wondering if it worth maybe dropping my tRFC down. I heard it can improve it a bit.

Yeah, you may be right about the VRM on your motherboard, but it's obviously not affecting it too much. Better cooling will do wonders for Ryzen, especially 2nd gen, and that may be what's holding your 2700X back the most. I ran my 2700X for a while with my Noctua NH-D15 before switching it back to the 7700K and my scores might be a little higher if I was still using it. It kept temps about 5-10°C cooler than the NH-U14S, which still does an admirable job (but that 7700K gets really hot, so I switched it back).

Side note - I had run R20 yesterday and scored exactly what you did for multi-core, but I don't remember what the single-core score was (and I forgot to check the box to save best score). Somehow, there was already a 2700X result showing up when I ran it both yesterday and this time, but I know that wasn't from me and I don't see it in anyone else's screenshot. I'm wondering if it somehow picked up my previous saved R15 score, converted it (somehow), and displayed it in the results. Strange.

Anyway, yes, I think it will definitely be worth it to tweak your memory. I mean, you have Samsung B-die and that's the fun of having it and half the reason it costs more. I'm definitely no memory expert. It's a lot of trial and error for me, mostly error. I went ahead and attached a screenshot of the memory timings using Ryzen Timing Checker.

 

[racer243l]

Back at it

5305 cb i9 9900k 5.0GHz cooled by H150i Pro 3600Mhz 32GB RAM on a Maximus XI Hero Wi-Fi

Almost got the 5.1Ghz scores... intresting for sure...

 

[ArbitraryAffection]

Yeah, you may be right about the VRM on your motherboard, but it's obviously not affecting it too much. Better cooling will do wonders for Ryzen, especially 2nd gen, and that may be what's holding your 2700X back the most. I ran my 2700X for a while with my Noctua NH-D15 before switching it back to the 7700K and my scores might be a little higher if I was still using it. It kept temps about 5-10°C cooler than the NH-U14S, which still does an admirable job (but that 7700K gets really hot, so I switched it back).

Side note - I had run R20 yesterday and scored exactly what you did for multi-core, but I don't remember what the single-core score was (and I forgot to check the box to save best score). Somehow, there was already a 2700X result showing up when I ran it both yesterday and this time, but I know that wasn't from me and I don't see it in anyone else's screenshot. I'm wondering if it somehow picked up my previous saved R15 score, converted it (somehow), and displayed it in the results. Strange.

Anyway, yes, I think it will definitely be worth it to tweak your memory. I mean, you have Samsung B-die and that's the fun of having it and half the reason it costs more. I'm definitely no memory expert. It's a lot of trial and error for me, mostly error. I went ahead and attached a screenshot of the memory timings using Ryzen Timing Checker.

View attachment 118317

Thanks! I might try and punch in 350 T or something in tRFC to try that. Crunching perf is great but want to get the most out of my system in games. btw my soc voltage is 1.08V currently and RAM 1.35. did you bump voltage to get these timings and speed or leave it at stock? thx

also might get a bigger noctua, do the larger 14 and 15 coolers use the same mounting equipment as my 12? so maybe i could literally swap out my current cooler and screw in the bigger one without swapping brackets. (I LOVE Securifirm 2) actually now that icheck i might not have enough room for the larger cooler, not unless i remove my wifi card

Spoiler: it's in the top slot

:/ sound card already in the bottom one. Didn't have a lot of luck with USB adapters either lol. Currently the 2700X hits 76*C, crunching with my "maximum turbo" profile in ryzen master. which is too warm for my liking (140-150W package power, ouch). Would be nice to have the 600-700Mhz increased clocks over my 3.4ghz @70W profile but doubling the power use and stress on my tiny, baby motherboard VRM (I think is a non doubled 3 phase, with just duped chokes and fets, but not 100% sure). I did want a new motherboard though. I mean i love this one but i could use a full atx one, and my case is full atx. I was looking at this one , what do you think? ideally it has to be really cheap haha. Sorry i typed a lot, bit hyper today^^

Back at it

5305 cb i9 9900k 5.0GHz cooled by H150i Pro 3600Mhz 32GB RAM on a Maximus XI Hero Wi-Fi View attachment 118319

Almost got the 5.1Ghz scores... intresting for sure...

You are alarmingly close to that 1950X result

EDIT @racer243l : Btw i worked out from our single core scores we have almost identical IPC and near linear clock scaling in this test. My result of 440 @ 4.3 GHz ia 16.8% lower than your result of 514 @ 5.0 GHz. You are clocked 16.2% higher. CB obviously not using vector widths much over 128-bits, or if they are, it is not fully saturated. Very nice result from Zen+ here IMO. What is interesting is that you have a better MP ratio. That is not something I expected given my theory on Zen having better SMT implementation (it really does).

 

[Arctucas]

Overclocked

Stock

 

[TWK_OCZ]

Pentium G5400 @ 3.7 GHz ( Stock Cooling )

 

[racer243l]

Thanks! I might try and punch in 350 T or something in tRFC to try that. Crunching perf is great but want to get the most out of my system in games. btw my soc voltage is 1.08V currently and RAM 1.35. did you bump voltage to get these timings and speed or leave it at stock? thx

You are alarmingly close to that 1950X result

EDIT @racer243l : Btw i worked out from our single core scores we have almost identical IPC and near linear clock scaling in this test. My result of 440 @ 4.3 GHz ia 16.8% lower than your result of 514 @ 5.0 GHz. You are clocked 16.2% higher. CB obviously not using vector widths much over 128-bits, or if they are, it is not fully saturated. Very nice result from Zen+ here IMO. What is interesting is that you have a better MP ratio. That is not something I expected given my theory on Zen having better SMT implementation (it really does).

Oh yeah, didn´t notice that. Strong showing from AMD right there.

Overclocked

Can you add CPU-Z and your cooler?

 

[ArbitraryAffection]

OverclockedView attachment 118321

StockView attachment 118324

wow that's a mental score for the 6700K overclocked. you are 5 Ghz~? Given 6000 and 7000 and 8000* and 9000 intel series all have same architecture it is easy to deduce the fact from other scores

*exclusing those mobile U series Cannon Lake parts of course.

Can you add CPU-Z and your cooler?

I assume is custom loop with D5 pump and PA140 triple 140mm radiator as per their specs. makes sense given the high score. he must also be delidded for sure, or super good chip^^

edit: @racer243l about the lower MP ratio on my 2700X than your 9900K, i brain farted XD it is because my single-thread clock speed is much higher than my all-core clock speed. Whereas i assume you are 5 Ghz single and all core as manual OC?. my single core clock rate is 4290MHz and my all-core is 4041MHz. so that explains the disparity and lower MP ratio for me . if all my cores were ticking at 4290 Mhz the ratio would be a lot higher for me, maybe allowing my 2700X to score a higher MP ratio than the 9900K due to finer granularity in FPU and wider core design for SMT. ^^

 

[xkm1948]

8700K 4.9GHz overclock at 1.248V
Corsair H115i AIO
Corsair RGB Pro DDR4-3200 32GB
Strix z370-ITX

 

[Final_Fighter]

r5 1600 @ 4.125ghz smt off, ddr4 3200mhz 16-17-17-17-34-54-1t, arctic liquid freezer 240. just posting single threaded score.

lets see if i can do 4.2ghz.


r5 1600 @ 4.2ghz smt off, ddr4 3200mhz 16-17-17-17-34-54-1t, arctic liquid freezer 240.

i dont no if i should push it further or not. still gots .5V i can add.

 

[ArbitraryAffection]

r5 1600 @ 4.125ghz smt off, ddr4 3200mhz 16-17-17-17-34-54-1t, arctic liquid freezer 240. just posting single threaded score.

View attachment 118357

lets see if i can do 4.2ghz.


r5 1600 @ 4.2ghz smt off, ddr4 3200mhz 16-17-17-17-34-54-1t, arctic liquid freezer 240.

View attachment 118362


i dont no if i should push it further or not. still gots .5V i can add.

If temps are ok, it should be fine. See if you can match or beat 440 single core. that is what my 2700X gets at near 4.3 ghz.

 

[Final_Fighter]

If temps are ok, it should be fine. See if you can match or beat 440 single core. that is what my 2700X gets at near 4.3 ghz.

ill see if its possible.

r5 1600 @ 4.3ghz smt off 2cores only, ddr4 3200mhz 16-17-17-17-34-54-1t, arctic liquid freezer 240.

ill see if 4.4ghz is possible like this but i think this could be it.

 

[LoLo2207]

My old i7 920 @3.8GHz (because since last week it can't go to 4.0 anymore )
Using a EKWB S360 loop + 1070 also with a EKWB block.

 

[SimpleTECH]

Tested all of my systems. The Atom CPU took a hell of a long time to run the benchmark. Thing is slow as hell.

2700X was custom water cooled and running at stock with PBO enabled. All of the other CPUs were also at stock and under air cooling.