Search the Community

So I made a function which benchmarks other functions. Because of my countless failures of achieving generics, I reluctantly had to make different bench functions for different data types. Right now, only the bool version is complete but I will add more functions. I have one question. I am using the same variable names in my union. In my function when I try to access "res", VScode Intellisense shows be something like "bool::res", which probably means it identified that it should use the "res" variable of bool type, because I am taking a bool return type out of the function pointer. But after I return the struct, in my main, if I try to access "res", then which "res" will it choose? I know I've done something wrong. #include <time.h> #include <stdbool.h> #include <stdarg.h> #define RUNS 5 #define AVG_RUNS 5 union result{ //to store the result of the function of any data type because GENERICS SUCKS. bool res; char res; short signed int res; signed int res; long signed int res; long long signed int res; short unsigned int res; unsigned int res; long unsigned int res; long long unsigned int res; char* res; }; struct bench{ // We do 5 runs, calculate average and repeat 5 times. Then we calculate the total average of all the 5 averages. We also maintain the highest and lowest. // Could have used arrays but who cares. long unsigned int run1; long unsigned int run2; long unsigned int run3; long unsigned int run4; long unsigned int run5; long unsigned int average1; long unsigned int average2; long unsigned int average3; long unsigned int average4; long unsigned int average5; long unsigned int total_average; long unsigned int highest; long unsigned int lowest; }; struct bench_result{ //Our bench function(s) will return this. long unsigned int elapsed; union result result; }; inline struct bench_result bench_bool(bool (*funcPtr)(va_list), ...) //Only works on boolean functions. Will add more functions which use other data types. All this because GENERICS SUCKS! { struct timespec start, end; //to measure time va_list args; //to pass the args to the function. va_start(args, funcPtr); struct bench bench; union result result; long unsigned int* runP = &bench.run1; //to iterate over the variables. long unsigned int* avgP = &bench.average1; long unsigned int avg_accumulator = 0ULL; //accumulator to add the values for calculating average for (unsigned char i = 0; i < AVG_RUNS; i++) { for (unsigned char j = 0; j < RUNS; j++) { clock_gettime(CLOCK_MONOTONIC, &start); result.res = funcPtr(args); //notice, we are storing the the result in our union clock_gettime(CLOCK_MONOTONIC, &end); *runP = end.tv_nsec - start.tv_nsec; //storing the elapsed time in the current run avg_accumulator += *runP; //adding the time to the accumulator runP++; //update pointer } *avgP = avg_accumulator / RUNS; //calculating average avg_accumulator = 0ULL; //resetting accumulator avgP++; //update pointer //long ass convuluted ternaries for calculating highest and lowest long unsigned int highest_current = (bench.run1 > bench.run2 ? (bench.run1 > bench.run3 ? (bench.run1 > bench.run4 ? (bench.run1 > bench.run5 ? bench.run1 : bench.run5) : (bench.run4 > bench.run5 ? bench.run4 : bench.run5)) : (bench.run3 > bench.run4 ? (bench.run3 > bench.run5 ? bench.run3 : bench.run5) : (bench.run4 > bench.run5 ? bench.run4 : bench.run5))) : (bench.run2 > bench.run3 ? (bench.run2 > bench.run4 ? (bench.run2 > bench.run5 ? bench.run2 : bench.run5) : (bench.run4 > bench.run5 ? bench.run4 : bench.run5)) : (bench.run3 > bench.run4 ? (bench.run3 > bench.run5 ? bench.run3 : bench.run5) : (bench.run4 > bench.run5 ? bench.run4 : bench.run5)))); bench.highest = highest_current > bench.highest ? highest_current : bench.highest; long unsigned int lowest_current = (bench.run1 < bench.run2 ? (bench.run1 < bench.run3 ? (bench.run1 < bench.run4 ? (bench.run1 < bench.run5 ? bench.run1 : bench.run5) : (bench.run4 < bench.run5 ? bench.run4 : bench.run5)) : (bench.run3 < bench.run4 ? (bench.run3 < bench.run5 ? bench.run3 : bench.run5) : (bench.run4 < bench.run5 ? bench.run4 : bench.run5))) : (bench.run2 < bench.run3 ? (bench.run2 < bench.run4 ? (bench.run2 < bench.run5 ? bench.run2 : bench.run5) : (bench.run4 < bench.run5 ? bench.run4 : bench.run5)) : (bench.run3 < bench.run4 ? (bench.run3 < bench.run5 ? bench.run3 : bench.run5) : (bench.run4 < bench.run5 ? bench.run4 : bench.run5)))); bench.lowest = lowest_current < bench.lowest ? lowest_current : bench.lowest; runP = &bench.run1; } bench.total_average = (bench.average1 + bench.average2 + bench.average3 + bench.average4 + bench.average5) / AVG_RUNS; //calculate the final avetage struct bench_result res; res.elapsed = bench.total_average; res.result = result; //passing the union itself va_end(args); return res; } Also, is it hard for you guys to read code on this forum? If so, I might post the code on some other website. It's shocking I don't have a GitHub account yet.

I can't find it... So I'm gonna casually assume it doesn't exist.

Hello, so I recently came across computer problems. I have since then done chkdsk, sfc /scannow, crystal disk and hd sentinel to check for my ssd health. I reinstalled windows and as for the ram, I have used passmark for my RAMS and it tested passed. I had a 240 gb boot drive and a 120gb for games, and I noticed that my pc had frequent mini freezes. I changed my boot drive using the 120gb one and the 240gb one for the game drive. The drives in crystaldisk shows that they are good and the tests like chk dsk and sfc/scannow have no problems but I still have the issue even with the new OS installed. The 240gb drive was in a 91% condition in crystaldisk but I want to make sure if it has problems, so, is there a software like passmark where it will test and diagnose the drive in an extended set amount of time? I'd prefer if it is a free software too! Thank you very much!

Hello, I was just told that Infinity fabric instability shows up as performance degradation, so I would like a benchmark (preferably fast) that will be able to convey this information.

Hi, So I'm building 2 computers, both with somewhat similar specs and I have a few questions. PC1 (My friend RGB Build) Part: Item CPU: AMD Ryzen 7 7800X3D CPU Cooler: Corsair iCUE H150i ELITE CAPELLIX XT Motherboard: Asus ROG STRIX B650E-F GAMING WIFI ATX AM5 Memory: Corsair Vengeance RGB 32 GB (2 x 16 GB) DDR5-6000 CL30 Storage: Samsung 990 Pro 2 TB M.2-2280 PCIe 4.0 Video Card: EVGA FTW3 ULTRA GAMING GeForce RTX 3080 10GB Case: Lian Li O11 Dynamic EVO ATX Mid Tower Power Supply: SeaSonic PRIME 850 W 80+ Gold Operating System: Microsoft Windows 10 Pro OEM Case Fan: Corsair iCUE AF120 RGB ELITE 65.57 CFM 120 mm Fans 3-Pack PC2 (My Dark Build) Part: Item CPU: AMD Ryzen 7 7800X3D CPU Cooler: Noctua NH-D15 Motherboard: Asus ROG STRIX X670E-E GAMING WIFI ATX AM5 Memory: G.Skill Trident Z5 Neo 64 GB (2 x 32 GB) DDR5-6000 CL30 Storage: Samsung 990 Pro 2 TB M.2-2280 PCIe 4.0 Video Card: Asus ROG STRIX GAMING OC GeForce RTX 4090 24 GB Case: Fractal Design North ATX Mid Tower Power Supply: SeaSonic PRIME PX-1300 1300 W 80+ Platinum Operating System: Microsoft Windows 10 Pro OEM Case Fan: Noctua A14 PWM 82.5 CFM 140 mm Case Fan: Noctua A14 PWM 82.5 CFM 140 mm The context: PC1 is for a friend, PC2 is for me. I play a lot of DCS World and Flight Simulator 2020 in VR. I care a lot about stability and longevity, I don't want my PC to crash or stutter or what not. Question 1: Is the Asus ROG STRIX X670E-E worth it versus the Asus ROG STRIX B650E-F for twice the price ? The X670E-E has 18+2 power stage while the B650E-F only has 12+2. From what I understand the X670E-E should be more stable. I don't really care about he I/O. Question 2: What tool can I use to benchmark the stability of the system under full GPU + CPU + RAM load ? Question 3: Should I go with Windows 11 ? I have an HDR monitor which I heard works better with 11, but I really hate the UI of Windows 11. But if it performs better than Windows 10 then I might consider it. Ah and it is possible to create a local only account ? I don't want to link my whole pc with my microsoft account. My Goal: I want to test if my friend computer is stable with my RAM and GPU before opening my motherboard box. If it is very stable I will return the X670E-E and get the much cheaper B650E-F. Thanks

I built this PC around a month ago and wanted to do a little testing. Specs: CPU: AMD Ryzen 7 7800X3D Cooler: Arctic Liquid Freezer II 360 GPU: Sapphire Nitro+ Radeon RX 7900 XTX MOBO: Gigabyte B650 Aorus Elite AX Memory: Gskill Flare X5 32GB (2x16) 6000MT/s CL30-38-38-96 (EXPO Enabled) PSU: MSI MPG A850G PCIE 5 (80+ Gold 850W) Disclaimer: I know very little about benchmarks, or what is or isn't a good score so I could just be an idiot, I'm just trying to make sure that everything is running properly before my return window closes for all my parts. I decided to download and run Cinebench 2024 and Unigine Heaven. I ran Cinebench first for the GPU and I got, what I believe to be a poor score? I don't know if being lower on the list than a 2070 super is bad...but it SOUNDS bad. These were my test results along with a screenshot of my afterburner near the end of the test. I ran everything stock and I haven't touched any settings for my hardware other than enabling EXPO on my memory. After I ran this test, I noticed I had an update for my GPU drivers in Adrenalin, so I downloaded that, restarted, and tried the run the test again, but now every time I try to run the GPU test, Cinebench just closes. I tried a couple times after a couple restarts and it keeps doing the same thing. I don't know if this is a big issue or not, but I didn't keep trying after that. Next I ran a multi core test for my CPU, with the following results. Again, I don't know if this is good or not but I felt it went ok? The stats on Ryzen Master were fairly constant through the whole test. I took a screenshot towards the middle of the test (shown) and again near the end, but the stats were nearly identical so I didn't post the second screenshot. Finally, I ran Unigine Heaven on both high and ultra. In the first Cinebench GPU test, the speed was around 3100MHz, and in the Unigine test they were fairly steady around 2800MHz. TLDR: I basically just want someone smarter than me to tell me if everything here looks ok, and if not, is there something I can do? I haven't really had any issues gaming so far, but I wasn't sure if there were any hidden symptoms here which might effect me down the road. Either way, I appreciate your time. Thank you

Post your human benchmark scores here! I'll rank people by who does the best in a certain test. No cheating! Link to website: http://humanbenchmark.com Here's mine: Best number memory: DECobra11 with 12 Best verbal memory: Fastest reaction time:

Noticed some very weird behavior of my laptop cpu after a clean install of windows. While benchmarking the cpu usage refuses to stabilize and power usage fluctuates between 12 and 17 watts even though the system has a 15w power limit. Intel xtu also shows the cpu power limit throttling while well below the 15 watts target. does anyone have a clue what could be the cause? Temperatures never exceeded 75°c during testing. lowering the power target yields the same fluctuating power and cpu usage. before the clean windows install it would be rock steady at 100% and 15 watts update: Increasing logging speed in intel xtu makes this even weirder. the cpu drops to 10watts for brief moments, only to then spike to 20 watts for a third of a second or so and then stay stable at 15 watts for a few seconds only to drow to 10 watts again. hwinfo shows the same tdp fluctuations update 2: this behavior only happens when the laptop is plugged in. When it is not plugged in it runs perfectly stable

I frequently check PassMark's Single Thread Performance list for new CPUs, because it is more important for games than multi thread performance. There is relatively new list named "Top Gaming CPU Performance": https://www.cpubenchmark.net/top-gaming-cpus.html Is this list reliable at all? Some interesting examples: CPU / Score 7900X3D / 9180 5600X3D / 7849 13600K / 6784 7950X / 5931 7900 / 3159

Hello, https://pcpartpicker.com/list/Kyqq6r I reccently upgraded my CPU from Ryzen 7 3700x to Ryzen 9 5950x. everything else in the system stayed the same. After installing the CPU i ran into a problem. When i had the 3700x, Cinebench had no issue at single and multicore test, however, with the 5950x, Multicore Cinebench benchmark turns off my screen while the PC fans are still crancked. seems like the Benchmark still on going but the GPU doesn't send any signal to the Monitor anymore. I have to turn the powersupply off in order to restart the PC. Some people say that i have RAM and CPU Compatibility issue. Can you guys please help me resolve this issue? Thanks,

Hello guys. Am new on this site and i realy need your help. I am a small streamer and i have some problem with my pc when i streaming. I make stream tiktok and yputube at the same time with low quality on tiktok and 1080p on youtube and some times i have streaming lag. I want from u to find some program to make benchmark when i have live to see where is the problem and what i need to change something from my pc. Thank u

Hi, I'm new to this forum. Not sure if this is the right area to ask this question. I'm using an ai benchmarker from this website, https://ai-benchmark.com/ranking_deeplearning.html. It was the only ai benchmarker I could find that I could understand. For some reason the results are in AMD's favor. I always thought ai wasn't good enough on AMD, but it's beating out the 4090! Am I using this benchmark wrong or has AMD really gotten good recently? Also if anyone knows another benchmark I could use, that would be wonderful.

Every 2-3 minutes my FPS seems to drop to about 10-30 FPS for a couple of seconds in every game i tested, warframe, eft and csgo. My graphics card and laptop drivers are up to date, tried installing windows 10, and now back to fresh windows 11 and still have the same issue. Yesterday, I changed my thermal paste with Arctic Mx-6, I had liquid thermal paste on cpu, but used normal thermal paste. My cpu is thermal throttling, but the performance is much better than before. I used the hole tube 2g of thermal paste, opened again my laptop and put more paste on cpu (on top of the new existing one, my cpu has a lot of thermal paste), but more on gpu I didn;t because I ran out of thermal paste. Tried stress testing gpu and cpu at the same time, but there I don't have the problem, nothing is happening, only in games. In games/stress testing cpu ang gpu temperature looks fine for a laptop. Specs: i7-10750h rtx 2060 24gb ram when its lagging when its not lagging 0712 (1)(3).mp4 0712 (1)(4).mp4 1.Should I buy another thermal paste to use on cpu ang gpu? (arctic mx4 or something more expensive like noctua)(arctic mx4 8 grams is cheap 5 euro) 2.or buy thermal paste on gpu and liquid thermal paste on cpu 3. can i use thermal paste on vram or i need thermal pads ( i used thermal paste, but before it had thermal pads) Edit: tests before opening laptop and using new thermal paste Edit: I disabled intel boost techn. and cpu, gpu temps are max 85 and 80 degree with 100% utilization. I dont think its lagging like that due to thermal throttling.

Am I just screwed? I bought my laptop from XoticPC (its just a clevo laptop, a lot of the time I see this laptop referred to as a Sager online) and these are the relevant specs: https://clevo-computer.com/en/laptops-configurator/geforce-rtx-40-series-clevo-laptops/geforce-rtx-4070-notebook-gpu/45103/clevo-np70sne-17.3-fhd-144hz-intel-i9-13900hx-nvidia-rtx-4070 13900HX 4070 (mobile) 64 gigs ram at 4800 MHz On a whim I downloaded 3DMark and ran the gpu test, hey look at that my gpu is in the top 45%. Whatever I got a little bit lucky but that's within the bounds of error. I run the cpu benchmark and my scores are absolute garbage. In the bottom 7%! The average score is like 12,000 overall and I was at 7,000. I updated windows and my nvidia drivers just as a sanity check and my scores did actually improve from 7,000 to 10,000 which is a pretty big bump but still really pretty low (bottom 15%). Did I just lose the silicon lottery? Is Xotic or Clevo doing something? I tried poking around the BIOS but couldn't find an XMP setting, it's a super weird looking BIOS (insyde, I included a screenshot with relevant info in it). I checked and saw that the BIOS was up to date with what they had on clevo-computer.com. Any ideas? I'm at a loss and I'm sad if I just happen to have a bad cpu.

Current status: GPU Arrived, stress test successful (Furmark Burn in 17 minute) This post will be continuously edited to update the GPU's status. Wish me luck So many of you probably aware LTT recently released a video about buying second hand 3080. Although I placed my order a week earlier I figure someone else will be intersted. I bought a 3080 mostly because of outside factors: A people close to me have their graphics card (GTX 670) failed (driver resource access error, DDU/repasted and still the same problem. Also the single fan not holding up, so suspecting the VRAM finally gave out, it didn't have thermal pads on top anyway). His PC still can boot into Windows using onboard (an i7-4770, the system is a Asus G10AC from my BF) Since I have a old GTX 970 why not upgrade? It was also a few days after 4060Ti 8GB released, I intended to upgrade to a 3070 years ago, but you all know about the crypto boom. Anyhow, since 4060Ti 8GB was such a disappointment, I was considering wait for a 4060TI 16GB. I wanted a 40-series because of DLSS 3, but then I looked at secondhand market, noticing a used 3080 is about the same as a NEW 4060Ti 8GB. Of course before that I already looked at other new GPUs like 4070, 6800XT. I wanted to buy a new card, but on the other hand I bought several secondhand cards in the past, most without any problem. I bought a EVGA 570 in US eBay, then the aforementioned GTX 970 from Chinese JD.com (refurbished with 1-year warranty from refurbisher) Following is a table I compiled to see which GPU is better for me. "Price" are lowest using price compare software in my market (Sweden, all price in SEK). "Price High" are lowest Gigabyte model or second lowest (I'm using an Aorus x570 elite, so I lean towards a Gigabyte card (one less RGB software to worry about), mostly for RGB I also used several power supply calculator to check if my PSU (Silverstone ET650-G) can take it. I watched several reviews and ranked the cards from left to right vaguely in relative performance, I was going to list other reviewer's results, but most reviewer only posted fps. I'm too lazy to calculate the relative performance. Ideally I want a 4070, and from this table evidently getting a 3080 with price of 4060TI 8G and performance of a 4070 is a very sweet deal. So 3080 it is, before putting down money, I spent an extra week tracking eBay bidding prices (add to watch list, then come back after list ended to see the final bid) 12GB model are had to find, but 10GB models are plenty. Although 10 GB is on the small end, I figure I can upgrade again in 2-3 years when 8GB or less really become a issue (right now only like 2-3 games can run on 4060Ti 8GB) All price above listed including shipping, in SEK, limited to Europe to avoid extra import taxes. I'm quite sad about did pull the trigger on an Aorus Master sold from France, most are Gigabyte Gaming OC, a decent card. Some people didn't know the difference and bid higher on a worse Eagle OC. (Gigabyte goes from Windforce OC -> Eagle OC -> Gaming OC -> Aorus Master -> Aorus Exterme/Aorus Waterforce) I tried to bid for some of Gaming OC cards, but at the same time I saw a "Buy it now/Best offer" only listing from Finland. The 3080 Aorus Master is also a rev 2 with better thermal pad. (yes, I looked up old review on 3080 Aorus Master) "Buy it now" price is about 450 Euro, higher than prices in the table. Although the listing isn't super attractive to me as the seller is selling 5 Aorus Master at the same time, a red flag for mining cards. But since Aorus Master have a quite big cooler, I decided to chance it. Plus other sellers on eBay often also sell many cards, meaning those can also been mining cards. The Aorus Master seller at least don't also listed a ASIC mining rig... Since seller also allowed Best offer, I sent in an offer for about 400 Euro, thinking might get a counteroffer at around 420-430 Euro which will be closer to the other prices I have seen. Surprise to me the seller accepted the offer, so I instantly paid the seller. Someone got the the MSI Ventrus OC when I'm waiting on my card, but still mine is second lowest. So now I'm a proud owner of a Gigabyte Aorus Master 3080 Rev 2 Pros and Cons I concluded for for this listing: Pros 1. Seller knows when the card was bought (June 2021), meaning the card is still in Warranty. (today is June 2023) 2. This is a higher end 3080, meaning even if it had been mined on, the beefier cooler should have kept the GPU temp under control. 3. Seller located in Finland, relatively cooler year-around. So even if it had been mined on, the card probably won't get baked. 4. It's and Aorus Master, with a screen on the side. We all know RGB automatically +5 fps, so probably automatically +20 fps? (<- This is a joke) 5. After payment Seller provided the original receipt. (hope I will never need that) Cons 1. Since seller listed/sold many GPU in the past. So it can be a mining card. 2. Not many pictures from seller so the card's exact condition is unknown. 3. Recent news emerged that there are Gigabyte 3080 cards cracking at PCIe slot. As an engineer myself I don't consider that is a QC problem, more like user-error on not using a GPU support bracket... (my grudge is confirmed once I have the Aorus Master in hand) The seller shipped the card in 2 days while the card is in transit, I asked the seller for the original receipt just in case I need RMA. Appears that they paid 1000 EURO for these cards new... and it also sort of confirmed my suspicions, no one but miners will buy 5 cards at same time. But I'm quite happy with what I got. Next up, unboxing and testing. *I will update this thread with posts about the GPU once I tested it.

I am looking to evaluate a bunch of different laptops for battery runtime with a focus on office productivity usage. Particularly I am after a benchmark that will simulate general office usage (working on documents, web browsing, very light multimedia etc) that will continue to run until the battery runs flat. A "high score" is not important but rather how long a laptop can sustain an office productivity use is what is import. A quick look around and PCMark 10 appears that it will fit the bill but wonder if there are any other options worth checking out?

HEllo good fellas from LTT forums, i have a simple question, how does your task manager graphs look like while rendering animations in blender, is this normal?, i know i know, i need more Ram, but the thing isnt thermal throtling, my room is like 15C.

Hello, my 3DMark score is 2000 points under average with the same hardware, that's 1/3 the score of my old pc. I feel like I'm missing out on performance. My graphics card is a Powercolor Fighter Radeon RX 6700 XT which is quite slim with only two fans. My CPU is a 5700x cooled by a Dark Rock 4. They're both not overheating. Any ideas why it is? Thanks for reading, have a nice day!

https://imgur.com/a/BZfyt6D I thought a 9700k @ 5ghz should be able to get 10k on multicore? Is there a setting ive forgotten in my bios or something? Im running my cache at 4.7ghz, 16gb 16-18-38 3200mhz corsair dominator memory 4x4gb

So, I got a new processor. Ryzen 5900x. Technically I RMA'd the first one cause it kept crashing at stock settings so I got two of them. The new one is... underperforming. not by a lot.... 5-10% below average as reported by several sites. So, knowing how sensitive ryzen is to ram I tried overclocking my ram a bit. Here's where the story gets.... weird. With tighter ram timings, and faster speeds, my cinebench r23 score went up. By about 4% from 20172 to 20933 BUT PassMark ram bench shows my latency has gotten dramatically WORSE which makes no sense..... the cas latency is the same (wouldn't drop to cas16 stable) the secondary's are slightly better, and the speed is up from 3600 to 3800 This violates everything I _thought_ I knew about ram and overclocking it. Also this motherboard will only set an even numbered cas latency which is weird to me coming from intel. my last platform (7820x) I was running ram at 3400cl15 Ram rated profile - 3600mhz cl18-22-22-22-42 1.35v Overclock profile - 3800mhz cl18-21-21-21-41 1.355v FTR the overclock passes a full 4 loops of the memtest86 tester. It's stable as it is.

Hey everyone, some time ago I notice my PC overperforming some days. Yes, it performs better some days than others in particular my GPU. I've been looking in forums for some reason for it to happen without success, no similar case. No updates or config changes between tests, just a restart. By the way, the performance changes upon restarting or putting my PC in a sleep state, that's all it takes. Specs: -Ryzen 7 5800X 4.65GHZ@1.235V -MSI B550 Tomahawk -2x8GB G.Skill Trident Z 3600MHz CL18 -Sapphire RX 6700 XT Nitro+ I tested my setup performance using FurMark (the first tool I used and got aware of this situation), NiceHash and Rise of the Tomb Rider. Radeon Software profile for NiceHash is 90% clock and 90% voltage The rest of the time I run my GPU at 100% clock speed and 96% voltage Expected Performance: FurMark Score: https://gpuscore.top/furmark/show.php?id=463848 Overperforming GPU: FurMark Score: https://gpuscore.top/furmark/show.php?id=463771 I ran both tests the same day, just a restart in between. I guess it happens 1 out of 5 system boots. Why is it happening?

Hi, Im working on overclocking my samsung bdie corsair vengeance pro rgb 3200mhz on GB Aorus Elite and 5600x. While testing, Aida64 fails almost immediately even on a slightest OC from 3200 to 3333mhz 1.4v and Vcore Soc 1.1v, nonetheless, passes on Prime95 RAM stress test for hours of run on even harshest OC 3400mhz, tighter timings and higher voltage 1.45v DRAM Voltage and 1.15v vcore SOC. Also it passes OCCT stress test as well. Should I ignore Aida64?

I undervolted my i5-8265u and ran all 4 versions of cinebench i know at the same time...like starting all 4 of them at the same time is a stresstest like this valid in any way?? this test lasted for 18 minutes in total and the scores were pathetic (of course)

Do you think the speed difference between Cloudflare and NextDNS is noticeable?