Is Cinebench Good
Gigí Ruais
Below is my average results running R23 Cinebench, even on PBO only, my Multi-core can score easily 11.9k, and if I boost it, Prime95 stable and stuffs we name it all, it could hit 12.4k Multi Core when letting all core runs at 4.69-4.7 ghz, and yes it's a very stable and achievable set up.
However, my Single core can only hit 1510-1540 stock to PBO, if I boost it max max, all this chip can give me on Single Core R23 cinebench would be 1590 average.. I don't know why it is like impossible for me to hit 1600 mark single core, while many people online shared their 1570-1590 single core even without really boosting, where I got only 1540 max. I know it is not a huge deal, basically only around 2-3 percent margin that I am chasing here.
16gb b-die 3800 cl16 (DOCP profile 4000mhz) but to get no WHEA error my best was 3933mhz with higher VSOC, so I decided to just run what possibly the max of most reliable and sustainable clock out of thie chip, FCLK 1900
Also I provide Timespy Extreme result, which is somewhat impressive on the CPU side when I compared to 5600x that runs at 4650 max, GPU not overclocked (undervolt tho), CPU same identical setting, no boost, PBO and pretty easy CO.
Sorry I see nothing in RM that indicates a problem. It is clear you are not running Cinebench which might have shown something. I think it will really be difficult to get the final few points out of Cinebench. Thanks and have fun, John.
Those temps are really good, my favorite CPU benchmark is Cinebench R15 personally. We also have a thread that you can post your scores on here:
Cinebench R15 cpu+overclocking score Overclocking I thought it would be cool to trow a new battle uphere. Try to overclock the hell out of your cpu, and post a screen of your Cinebench R15 scores. And share your thoughts. Tools: - Cinebench can be found here: - CPU-Z can be found here: -z.html Rules: - Your printscreen must be clear, No wierd censorship - Your printscreen must be included CPU-Z so that we know your clock speeds. - Your screen must be from your self, so dont google for amazing fake scorings. - Keep it cool, no flame wars. Lets start the battle! DemonX09s Scoring Spreadsheet, to view results. -Jpdpom-EODneLaJhDcecT5I8rG1wI/edit?pli=1#gid=0
Handbrake is a great tool in general and depending on codec and settings it can utilize a lot of cores. Asus Realbench has it as a component integrated along with gimp image manipulation and some other open source stuff. I like Realbench or Aida 64 for stress testing and Cinebench R15 for a general comparison of computational power.
Benchmarking or stress testing?
Handbrake, or rather x264, is good but for stress testing I'd rather use something else, like RealBench which was already mentioned above or the x264 Stress Test v2.
x265 is also a pretty good choice as it hammers down on your CPU really hard. You could for example use HWBot's x265 Benchmark. Using the overkill mode multiplier for 4K encoding is going to give you a pretty damn long stress test.
However if you want to compare your results to something else, eg benchmark, then use Cinebench R15 or the HWBot x265 benchmark or something similar. They're quite widely used and as such it's easy to compare.
When in BIOS, all power management features of the processor are disabled. In addition, all interrupts are disabled. One core will be running in a tight polling loop checking for changes and input at all times. Temperatures seen in this state will thus be higher than those seen in any runtime (Windows, Linux, etc.) environments, even at idle.
i dont know whether this is an issue with the chip or what but im using a 240mm aio aswell and the temps just dont feel right, whenever im playing games its running at 87c ish and throttling to 4.7ghz the game i was playing was farcry 6 which is quite heavy but mainly on the gpu so i didnt expect it. I NEED Advice ASAP.
my cpu just chills around 2-4% usage rarely seeing 5. i use the artic mx-6 and voided my motherboards warranty with a contact frame. using a guide for the perfect amount of paste aswell as the waterblock is mounted right my temps are just not making sense at this point.
To be honest, if that was my problem, i would disassemble the water cooling and clean everything i can, and then with open loop i would stick one hose into some bottle and see if the pump is pumping.
If the water cooling works, i would undo the frame, and use the stock bracket, then reassemble everything and see what temps i get. From my point of view, there is either bad heat transfer CPU->Water block, or the water flow is just to little....for example, if the liquid gone bad and some plancton lives in the water block.
What happens if you turn off the pump while PC is on idle? the temp should gradually rise, but also not extremely fast, i would say it could take like 30 seconds mayne 1 minute to reach 90*C, at which point i would turn on the pump....the temp now should spike down to some 30*C within like less than 5seconds.
If the temp woth pump off rises very quickly, then CPU has little contact with the water block. With pump back on and temp dropping very slow, it would suggest, there is a problem with the water cooling.
Hi i recently bought the i7 13700k , i went for DeepCool LS 720 for the AIO, big wented case, i get in idl/browsing minor load 33-38 C , gaming 45-62 C ,even RDR2 after hours of gamplay max settings wont hit higher than 65 C, But when i do Cinebench R23 temps are 100c, Any aplication that uses load over 75 % on cores i get temps over 90 even up to 100C should i be worryed??? never had a CPU this hot before , i do not overclock it and it is not Undervolted eather, just auto settings in BIOS.
I downloaded Last of US part I and it does a shader build in the begining, well till that is done CPU gets 92-95 C? is this normal? i have to mention that shadaer build takes over 10-20 mins , is it bad for CPU to run that long that hot???????
size of the radiator in this case means nothing, bcoz the die of the CPU is tiny considering it does 250W.
I got here twin EK-CoolStream 420 and even with fans on full power the CPU hits 100*C....that is bcoz the water block on the CPU can not transfer enough heat onto the water.
First of all thx for ur answer,and unfortunatly, i have an even cheaper motherboard, gigabyte 760, so im not even considering overclocking, and i have to say that this is my first Water cooler, and at first i thought that it is broken cause when i saw cpu temps at 100C and i touched the radiator ... that was cold... not even a bit warm... i expected that at these heat levels, at least the radiator will get a bit warm but no mine is cold. is this normal? (only by touching it never used mesuring equipement ), or it is explained by the fact that the die size is to small and block cant transfer enough heat? or i did somthing wronG?
i mostly game on it, and sometiems some video editing, so temps there where ok. but im realy curious how do ppl cool this thing, i saw posts on forums that whit cinebench r23 and no undervolt they never passed 93 c mark... how is that possible?
Intel does not verify all solutions, including but not limited to any file transfers that may appear in this community. Accordingly, Intel disclaims all express and implied warranties, including without limitation, the implied warranties of merchantability, fitness for a particular purpose, and non-infringement, as well as any warranty arising from course of performance, course of dealing, or usage in trade.
Breaking down the instruction lengths shows a tail of longer instructions, with few instructions reaching 8 bytes or longer. Instructions that are 4 bytes or below are likely scalar integer instructions, while longer ones would be AVX ones. Cinebench 2024 contrasts with Y-Cruncher, where heavy AVX-512 use skews the distribution towards longer instruction lengths. 7-Zip is the opposite. Its exclusive use of scalar integer code gives it shorter instruction lengths on average.
Plotting all of the BTB sizes across tested CPUs gives us a good picture of how many branch targets we have to cache for Cinebench 2024. 512 entries is not enough, but a 1536 entry BTB gets you most of the way there. Passing 7K BTB entries will almost eliminate decoder redirects, or BAClears in Intel terminology.
On the Ryzen 7950X3D, L2 hit bandwidth can exceed 350 GB/s. Code fetches account for a substantial portion of that traffic. Combined with data-side accesses, the L2 gets a request roughly once every ten cycles.
Thus, Cinebench 2024 rewards CPUs with high scheduling capacity for integer and FP operations. Assuming enough scheduler capacity is available, Cinebench then benefits from larger load/store queues and ROB capacity.
Execution units on a CPU are organized into different sub-pipelines, fed by ports from the scheduler and register file. Instructions that go to different ports/pipes can start execution in the same cycle. CPU designers try to organize execution units to spread load across different pipes, and try to avoid situations where IPC is limited because too many instructions have to go down a small subset of pipes.
By itself, Cinebench 2024 is a moderate IPC benchmark with a sizeable instruction and data footprint. Code spills into L2, but the instruction stream is easier to follow than what we saw in games. Decoupled branch predictors can thus keep the frontend fed even in the face of L1i misses. On the data side, Cinebench 2024 spills out of L3 and requires a modest amount of DRAM bandwidth. High scheduler capacity across integer and FP operations help keep more memory operations in flight in the face of DRAM latency. In that sense, Cinebench 2024 can be seen as Cinebench R23 with more emphasis on DRAM performance. When hitting the execution units, Cinebench 2024 uses scalar and 128-bit packed floating point operations. Wider vector execution units are not useful. Scalar integer performance plays an important role in keeping the FP execution units fed.
If you like our articles and journalism, and you want to support us in our endeavors, then consider heading over to our Patreon or our PayPal if you want to toss a few bucks our way. If you would like to talk with the Chips and Cheese staff and the people behind the scenes, then consider joining our Discord.
c80f0f1006