valzo

@bardozVAL @IgorWallossek preach

sorry but this article is very hard to take seriously. I am sure you have the best intentions but the testing methodology itself presents some very obvious issues and you make some claims that are unsubstantiated by your sources. 1. first of all, if your LDAT is measuring >20ms e2e, something else in your pc is being a massive bottleneck, making your setup not representative of the people that would benefit from increasing their hz. 20ms e2e is absurd - you'd have single digit (<6ms) e2e on any competitive modern FPS, which already changes your whole point of "it's a small percentage of the total". Specifically, with my LDAT, I have <3ms (almost always <2ms) e2e on a dx11 app that renders a pure black image and flashes a white image on input. So for you to be getting 20ms it means you're testing with your FPS locked to a very low value or with something else going very wrong in your pc. 2. You test latency based on clicks, but you don't verify whether "mouse 2" uses 8khz for clicks no matter what the reporting rate is set to, which is very common. In which case, you basically thought you were changing a variable while you weren't. This is a very big oversight. 3. n=25 is incredibly low for this. I only start getting serious convergence after 500 clicks, because you need to consider that the various phases of the quantization steps (usb polling, monitor refresh rate, fps) aren't phase locked and slowly drift, and you want to be able to sample for long enough that you don't let the current relative position of the phases dominate the testing. So even if 1/2 were solved, you can't run n=25 and make claims based on that. 4. you only consider the latency effects of higher hz without considering the fact that increasing reporting rate increases the GRANULARITY of the data, the decrease in the upper bound of latency per packet is only really a side effect. Increasing the granularity of data has effects on motion blur by means of decreasing jitter in camera rotations due to making sure that any motion being reported by the mouse is evenly spread out across frames - higher granularity decreases the likelihood of a frame being simulated with no motion packets during a continuous mouse movement. This is impactful on motion blur by means of the temporal integration occurring in your eyes, but ofc I'll summon the expert @BlurBusters on this topic, but the effect of higher hz on this phenomenon is very easy to reproduce and I've done so myself both on the mouse cursor and in-game by properly adjusting the shutter speed of my global shutter camera 5. you make some claims about raw input that are not substantiated in your sources. "Windows explicitly documents in the Raw Input model that even with 1,000 Hz mice, multiple input events can accumulate between two iterations of the message loop" -> you are missing the point. This isn't an issue, it's intended behavior. By the time you read your message queue, it's completely normal that, since this is all asynchronous, more rawinput messages could have been added to the queue. The way you use this fact is a non sequitur. You say: "That is a fairly clear indication that high input frequencies are not automatically translated 1:1 into immediate benefit". This does not follow logically at all. Modern FPS competitive games (cs2/valorant for example) use GetRawInputBuffer which consumes the queue completely right when the game is about to simulate its world (read: use the rawinput data to rotate the camera/execute shooting events etc.). So you are taking a rather "obvious" point (the fact that by the time your win32 app gets the lock for the message queue and reads the WM_INPUT message, more messages might have arived), to mean that those messages are inherently delayed, whereas any good implementation of rawinput would still consume ALL the available messages at the time of the game simulation by means of GetRawInputBuffer. So all this said, your conclusions at the end of the article are just not substantiated by the contents of the article. I am not saying 8khz is meaningful, mind you. I am saying that you are trying to make the logical leap that it is NOT meaningful by means of data points that do not, in fact, prove it not to be.

An Analysis of 8K Polling Rates: A Lot of Hype for Little Added Value, and Why Only End-to-End Really Matters in the End. Read more: igorslab.de/jgo9

@shwamper5 would be cool if you made it so the background of each skate would have a color to signify their speed

After a lot of work I'm finally happy to share my skate tier list that I have been working on for a while. After some critique from my last one I have made it a google doc with pages for descriptions/my thoughts on particular skates. docs.google.com/spreadsheets/d…

@HuskyLivin @Madness727 tripping

This is what a 149000k looks like after being mounted by the Arctic LF3 Contact Frame for about a year Completely bent CPU It's the worst mounting mechanism ever made and I don't know why Arctic won't change it Just use the old mounting mechanism, it's fine and simple

@jessickyeah let it burn your house down and then blame nvidia

wtf do i do

@aschilling why sell 9950X3D2 when you could sell 2x 9850X3D

So what about the Ryzen 9 9950X3D2?

@unikoshardware Finally… :D

karhu ram test v2.0 jan 14th 2026 cant wait! you should buy it now before the price increases #introduction" target="_blank" rel="nofollow noopener">karhusoftware.com/ramtest/#intro…

@itsHemu2K yeah thats fair point 😹 8400 would be sick :o

@1valzo Nah I only umc clocks 2000/2100/2200 etc. 8400 seemed to run for a bit but needs more work. Haven't even touched advanced settings yet tbf

c32 isnt stable due to temps

@itsHemu2K 8200 CL34 looks cool 😛

@1valzo 8200 but I don't like the number

@itsHemu2K whats the max frequency you can do?

@1valzo Yeah

@pursuitedx wow…

@facegamefps @Madness727 link would be awesome

@Madness727 You can buy new A die China 400$ cl26 6400

LMFAO NAH That's Crazy Ebay listing: 157526812901

@itsHemu2K 🔥

Black Ops 7 Optimization Results 📈 Setup: i9-13900K | RTX 4080 Super | 48GB DDR5 Before: 246 AVG | 181 Low 1% After: 362 AVG | 242 Low 1% What we optimized: ✅ BIOS configuration for 13900K ✅ DDR5 timings and frequency ✅ Proper GPU Overclocking ✅ Windows & config tweaks +47% FPS increase. Same PC. Better optimization

@jakeufps @Aerobrake77 wccftech.com/samsung-report…

@Aerobrake77 Wonder if we'll ever see a dedicated 24.5" OLED at this point, probably not. But hey, if AOC can make a literal CS2 monitor, maybe there's a niche for it.

Are there even any interesting products left releasing this year? Only thing I can think of is the 60HE V2 😭

@respawntalks @blazethew4y @itsHemu2K 🤦🏻‍♂️

@blazethew4y @itsHemu2K the way you know more than a self proclaimed overclocker 😭

If your PBO settings don't look like this, you're leaving performance on the table. Most people run stock or copy random settings. Here's what actually maximizes Ryzen performance 👇

🚨Crypto will collapse in 2 years By 2027, quantum computers will hack any wallet TRILLIONS in BTC, ETH and stablecoins will be stolen Here’s what happens next and how to protect your crypto👇🧵

@WallhackHQ @decode_000 take this

Nvidia GeForce RTX 5090 Giveaway🏆 Need a new graphics card? We’re giving away a Nvidia GeForce RTX 5090 to one lucky winner! Steps to enter: - Follow @WALLHACKHQ - Like + RT post - Tag one friend International Giveaway Ends October 31st ※ You can also participate via Instagram!

@PrayFPSHub @LookitsAllSkill 😭😩

Hey @LookitsAllSkill we doin payment plans or what?

@itsHemu2K 265K 🙂 OCN is already full of 285K testing 😴

285K or 265K🤔

@HyggeFN_ @Dety0 clearly user error

@Dety0 Det er fucking forfærdeligt, hvorfor fanden virker det ikke bare når man betaler 20k for en pc

Need someone to fix my pc I am getting the craziest spikees where my frames drops to 100fps every 10 seconds