Such insanely high quality content. If TH-cam’s algorithm doesn’t push videos like this into user’s recommended sections the TH-cam algorithm is trash.
Since some seem to have missed it. At 11:58 I make clear that I do *NOT* recommend to just use the maximum DPI value! The ideal DPI value depends on many factors like build quality of your mouse, quality of the sensor, firmware, etc. By simply choosing the maximum DPI value you _might_ make matters worse _if_ your mouse cannot handle that without issues _(jitter, increased delay, etc.)_ . The point I made there was that very *low* DPI values should be avoided - but don't just go to the other extreme. 😅 Also there will be a followup video where I plan to test more DPI values as well as wired vs. wireless. Not sure about the format of that video or video(s) depending on how many mice I will test.
Great video. So 800dpi will be same as 1600dpi? What will you recommended the dpi for Razer Deathadder v2 as afaik it has the latest best sensor may be! Thanks .
For the people wondering, your 'optimal' dpi depends on the sensor on your mouse, for most 'gaming' sensors the sensor will start smothing/add correction to your imput when it is over 2000 dpi so you should probably stay below that.
oh my god the quality dude you're freaking amazing I'm sure I'm gonna return here like 5 years in the future and I'll say hey I was here 5 years ago, that's my comment over there
as someone who used to play csgo on very low spec laptop, i used to run 250hz on my mouse because it lessens the load on the processor. The processor was so average that when i move the mouse on cs at 1000hz i lost ~10-20 frames (from around 90 to 70-80). By setting it to 250hz, i'm able to maintain 60fps on competitive matches, and no longer lose frames when moving my mouse. I now use a much better pc, but i probably still wont use 8000hz polling rate even when it's a standard. Just to save on processor power. Also this is an very awesome video, great tests and results!
@@silentplayz6797 ahhh, most old laptops = non replaceable CPU. and often people get them as hand me downs as they often do not have the cash to buy one. so.....
@@omarcere9383 today thats mainly true, but back in the day (in the core 2 duo days) many laptops had a CPU socket, the only thing you couldn't upgrade was the GPU.
@@yoyoninjaboy6431 That's kind of what latency means, doesn't it? Like with a Bluetooth mouse, it's an initial movement delay of 8ms over wired mouse, but beyond that the delay doesn't change.
@@filipbunalti yeah I guess I just wanted to state that it would only be relevant if moving a single pixel. Wouldn't it also only be relevant if you reached a breakpoint to process the input on an earlier frame?
@@yoyoninjaboy6431 I'd just like to know if it would make a big difference for me as i have been playing with 800DPI for a long time. Unfortunately Chris didnt show any tests with my exact DPI number so i can only assume a number between 400DPI and 1600DPI
@@yoyoninjaboy6431 Everything you do in terms of mouse motion would be "ahead" by that amount of time i.e. each frame of the game will have more "advanced" mouse position consistently.
Best viper 8k video I have seen you properly tested it, every other channle that say they test it dont they just say whats on the Razer website and show nice pictures. I have the Viper 8k and I have to move the mouse extreemly fast to hit 7600Hz im very unlickly to hit that speed so the mouse only averages around 4 to 5 k seams like a gimic to me, sure its smooth but so is a good sensor running at 1k Hz.
As an EE, I can appreciate the incredible amount of time/energy put into this. You don't put 300+ hours into a single video unless you have a true passion.
Insane testing, especially if you can automate it. I think you should run a website providing this information for as many mice as possible, I think you would be the only source for this information then.
The problem is that a lot of games have bad sensitivity settings. And when they are bad it's almost always better to have lower dpi because the sensitivity always is too high with high dpi settings.
@@rincx329 Well, maybe that is the case with older games, but any title released after 2015 should do perfectly fine with 1600 DPI and a lower in-game sens.
@@rincx329 maybe we can change the windows sens? mouse 1600dpi windows sens maybe 4 ticks or 5 instead of the default 6. Now the real question then again is would that introduce input lag since i guess the cpu would have to scale the input down so i guess multiply some shit depending on how the scaling is done. Idk how exactly input is processed
@@freesS3 You get less fine-grain control the higher your mouse DPI is, many games give you an arbitrary whole number scale of 1-20 or something like that. For example a game where I play with 400 DPI and '3' in game, and it only has full number steps, if I were to switch to 1600 DPI to maintain the same eDPI I'd want to switch my in game sensitivity to '0.75', which wouldn't be possible.
@@domzaeWoW Well, all the games I play will work perfectly fine for me. I play 30cm/360, so its not the lowest sens. I can easily half my ingame sens in any game I play. And I wouldn't even know any game which wouldn't allow me to half it either.
It wont because it is flawed. This is an Ad for the razer mouse. While its known that the input delay from higher DPI's is less they are generally speaking less accurate and thats why pro's use lower DPI's. High DPI usually introduces line skipping issues making it overall less accurate.
The reason people use lower DPI is that it makes your movement smoother and less jittery. Probably due to higher latency. I personally can track better when my mouse is set to 500Hz even lol. It just makes my movement smoother so I can track easier.
4:24 Because operating at higher poll rates can increase CPU usage and may delay other input chains (like your keyboard), or delay other more important processing chains within your PC. Having a real 1000hz poll rate is already excessive since your framerate and monitor delay is the main bottleneck to focus on, not your mouse.
i think there's a misunderstanding here. what you explained is a perfectly valid reason to slow down the rate of move events, but b(n)s was asking why one would slow down click events. the key difference is that the sensor can (and will) actually generate 8k move events per second when asked to do so, while the mouse buttons will only generate 1 event per click/release. so unless you perform 8k mouse clicks per second (which is physically impossible) the buttons will never overload the CPU with events and thus don't have to be restricted to a lower polling rate. so we actually have 2 polling rates now: 1: the actual USB polling rate (determined by the USB port generation) 2: the sensor update rate (determined by the user setting) except for the G203, in which case the user setting actually lowers #1
@@LastExceed But that's the problem. There's no way to individually set the poll rates for these parameters. You only have one setting within the driver panel (if you're lucky), and that might not even do anything depending on the mouse and its drivers. Nor is there a way for anyone to reasonably verify it, so manufacturers could blatently lie to you. Which is why I recommend to just set the poll rate to 500 or 1000 and not over, even if the mouse has the capability, since you're trying to get rid of the actual bottlenecks in your system instead of contributing to it. Since USB devices function on a system-interrupt basis, it really is a trade-off for higher poll rates which can have an impact on other areas of system responsiveness (even framerate in some cases), in exchange for such a minor improvement in mouse responsiveness in realworld scenarios. Having better frametimes and more consistent FPS is more productive to the overall end-to-end latency when you're already at 500 or 1000hz poll rate.
@@TheKillerZmile yes, because input handling is usually single-threaded. Also the actual CPU time requirement of input logic varies vastly from game to game. If its simple and well optimized then it wont be a problem even on mediocre hardware, but in some games even top tier CPUs will suffer
@@TheKillerZmile The amount of cores you have only matters as much as how well modern game engines can utilize them. We are still stuck in a transitional period where anything beyond 4 or 8 physical/virtual cores is not utilized well and single-threaded performance is still king for gaming. But even if you could run all IO related tasks like USB processing on a separate core/thread to the game or application, it would still take CPU time to process, and also time to communicate with the game engine.
I love having the "sharpshooter" button set to an extremely high DPI. It allows for fun things like rapidly spinning or ease of use with controlling certain aspects of games that force your mouse movement slower, like tank turret rotation or helicopters.
For future mouse movement tests, consider strapping the mouse onto a cartesian 3d printer. You should be able to easily automate the movement of the main plate relative to the mouse in 3 dimensions. No need to reset it manually. Probably fast and accurate enough for these tests too.
I considered that. Following problems: - how to ensure that the contacts touch - movement too slow - expensive What such a setup (i.e. drawing robot) would be great for is to test sensor accuracy.
@@BattleNonSense What about using a turn table for vinyl records? That would give consistent speed, but the vector would always be the same which is maybe not what you want. Also one must make sure to always put the sensor on the same "track" since speed will vary.
Chris, in your main conclusion you say that a low dpi value (like 400 dpi, which I believe it's the most popular) should be avoided bc your data shows it increases input latency. However, the values you show for the jump from 400 dpi to 1600 dpi are ~ 2 ms for 1000 Hz (again, the most common case) and ~ 4 ms for 8000 Hz, which don't seem all that significant, is this *small reduction (~ 2ms)* worth all the hustle of re-configuring and re-adjusting all the setting of the games and ones cognition? It doesn't seem so to me. Do I missing something? GREAT WORK! As an engineer you're a great inspiration.
Hey man nice video but why did you not check 800 DPI? I think 400 DPI is known to produce input lag while 800 and anything above should be fine ? im not 100% sure on that but i would love to see an update else i would need to go to 1600 DPI from now on haha
@@BattleNonSense nice to hear- thanks und dankeschöööön 🤗 / bin sehr interessiert zwischen 800 und 1600 dpi da ich meine razer im Treiber immer auf 800 stehen habe und dann ingame die weiteren Einstellungen vornehme - dank dir habe ich alle g sync Einstellungen + vsync und auch den afterburner auf meinem pc und erstmals spiele ich mit einer gleichmäßigen framtime und der Unterschied ist Gold wert 😘😘😘
I always felt 1600 dpi was simply better than 400/800 but this pretty much confirms it. More responsive and smoother. The only reason I use 800 is because it's better for the desktop. Now I'm going to try to switch permanently to 1600 for every game that supports it.
@@Rainnnny yeah just change your sensitivity that youve probably been playing on for years. its not that hard bro i've played on 400dpi for over 7 years and its impossible for me to go up
Incredible video. Here is some advise to gamers: don’t fall for the hype of ultra-high mouse polling rates (2000/4000/8000 Hz). It’s just the new marketable number, since no one cares about DPI anymore. Stick to 1000Hz or even 500Hz to save CPU cycles and mouse battery. It won't affect your gameplay at all. If you really want to reduce your input lag, improve your reaction time with a good sleeping schedule, good diet and exercise. Healthy habits can have a substantial impact in reducing your reaction time.
What a fantastic, thorough investigation! Thank you. Can I use the pointer speed setting in Windows to get normal cursor speed on the desktop despite using a higher DPI, or are there any issues with that setting?
Windows cursor speed other than the neutral setting introduces mouse acceleration on the desktop and in any software that doesn't use raw input (shouldn't be a problem for most shooters).
@@crownofarnias I have not seen changing the pointer speed in windows introduce mouse acceleration - that is with "enhance pointer precision" *disabled* ofc. :)
Looks like I was misinformed and the reason people recommend 6/11 mouse speed has something to do with skipping counts/pixels, not acceleration. Not sure how valid that is. Either way, the setting shouldn't affect the aiming in most modern shooters because of raw input. It might affect the parts of the game where you have a cursor though.
Battlenonsense were the mouse pad pushed at the same speed with the two different dpi values? If so the 1600dpi would be moving 4 times as much right? Could that be the reason for the lower latency? Also were you using usb 2.0 or usb 3.+? Also very curious to know if stuff like the systemresponsive setting in the registry database changes anything for your results? Can't shake the feeling that something like drivers from different mice was the cause of the polling rate not working properly?
I think lower dpi has lower latency because the sensor needs to travel 4 times less distance to send information to the computer that the louse loved but i didn't expect ut could have that much impact
USB 3.0 (aka SuperSpeed) is completely irrelevant for input device latency. The fastest interrupt service interval of Super Speed is 8000 Hz, the same as High Speed USB.
@@Rezler_ i am sure you mean higher dpi and according to my own test it is opposite of what you said.if i move my mouse slowly across the screen the polling rate is very low... somewhere around 15-25hz... if i start moving it faster across the screen it will go up to 1000 hz. It all depended on how faster i was moving the pointer on the screen not the mouse. So if you move a mouse physicly the same distance at the same speed the dpi will play a factor to the polling rate making higher dpi give the result of higher polling rate
Some people like to lower their polling rate because it interferes with USB power and can cause bugs in some programs. This happened to AntVenom in Minecraft.
@Battle(non)sense your testing/engineering skills is fucking insane ) I predict that nvidia will use your findings in new feature in RTX 40x0, as they did with your latency work in 30x0 ))
Very great video as always, you do so much tests and it's always the right metodology, I think lower dpi has lower latency because the sensor needs to travel less distance to send information to the computer that the louse loved but i didn't expect ut could have that much impact
awesome Video and awesome German Quality Engineering with the Idea on how to actually "measure" something that brings Data that is comparable. I would have loved to see the CPU Usage with 8k vs 1k and 500hz because if Razer did no special "Magic" with that 8k processing, the CPU Load should be huge.
I think people would use higher dpi if games wont spin like aimbot when you have lowest sensitivity and 800 dpi. I also wish every game has a sensitivity slider like csgo where you can enter any value like 0.85 without stepping over it.
It's so dumb. Even FPS game devs think people want to play 5 cm per 360. If a game has 50 sensitivity notches to choose from, 5 to 50 notches of the ingame sensitivity are uselessly fast. Leaving you with 5 values to choose from between 1-5 so you're forced to change your DPI to fine-tune.
Having 2 little kids increased my input lag tremendously :P When I think about moving my mouse it takes about 1 day to 2 weeks for it actually moves. But then it is mostly fine for a few minutes.
@@moonmessiah9368 Lol :) Nonono the pretty lights on the keyboard draw the 2 year old in...they have to be pressed/slapped and turned off :) And when they are off they have to be turned on and off and on and off :P Before I would get some time during naptime but that is over..no more naps .
Bought mine and every game works with the 8khz. A lot of people think it doesn't work right and for some it doesn't work. But in the end of 2021 it is worth getting!
Wow, amazing work! But... Wait what? Did Battle(non)sense just proved that 1600 dpi is better than 800 and 400 dpi? Sadly there was no 800 dpi tests which I got used to. Now Im confused and Idk I need to switch to 1600 dpi or 800 dpi is fine? And im talking here about Hero Sensor with 1k Hz (g pro superlight and my g305). @Battle(non)sense - can u explain to us if 1600 dpi matters only for 8khz? or we should also use higher dpi like at least 1600 for 1khz mouses? And im talking here about 240hz/1080p (with no more than 240 fps lets say). I think 800 dpi is perfect spot because even if I devined my sensitivity in game I feel my mouse is overshooting.
@@BattleNonSense after my subjective tests I think 1600 dpi can matters on 360hz only. For 1080p 240hz 800dpi is still enough. Still I wonder how pros can play on 400dpi and still be good
Lots of people like to use 400 dpi for general windows/application use -- do you expect something like RawAccel (mouse-acceleration programs operating at kernel level) or using something like 4/11 instead of 6/11 in the windows mouse speed settings to effect the input delay when switching to a higher dpi like 1600? If so, to what extent, would it be totally counter productive and eliminate any input lag gains from the switch from 400dpi to 1600dpi? I suppose a qualitative answer is the best to hope for until such a test is performed, but I'm curious to hear your thoughts!
Use 400-800 in Windows and cursor-heavy games and switch to higher DPI for shooters. Changing the settings in Windows itself, unless you're sure the game uses raw input, is asking for trouble.
Keep it up man, This is great work. A video I would love to see is windows settings and the influence it has on the gaming experience. Like windows game bar etc.
So a lower DPI causes an increase in delay? How interesting! This is significant considering the debate between using low DPI/high sens or high DPI/low sens..... I've always been told it doesn't matter which way you go since they "balance out" either way.
Multiple factors could play into this, but a higher DPI will obviously give you better movement resolution over time resulting in less latency. I didn't crunch the numbers, but depending on how fast you actually move the mouse this may or may not have a lot of impact. Further delay may also be added by data processing from the sensor which may depend on the mouse settings. In the end it's always an "it depends".
What is the impact of high polling when playing at more reasonable framerates (100-200)? This analysis is really impressive but it still doesn't explain what this tech will (or won't) do for most PC gamers. Until 360 hz and up monitors become the norm and GPUs are widely available to put out these kinds of framerates AND games are developed to run at these framerates without glitches and physics issues this all still seems to be creating a solution for a problem that doesn't yet exist.
8000Hz is really meant for those who already maxed out everything else. But if the mouse you want to buy anyway supports more than 1000Hz then it won't do any harm.
Totally get that and I'm glad they added it to the mouse rather than make it a "premium" feature. Just hope this doesn't turn into polling rate wars between manufacturers like with dpi and sensors in the past. Amazing analysis and quality AS ALWAYS, would just like to see more of what the impact is for more typical fps and any insight content creators might have from razer regarding support for this tech in games going forward.
@@yoyoninjaboy6431 Well, there is a cost in multiple regards: You need a USB High Speed PHY, which is generally more expensive (probably negligible in relation to "gamer" mice prices) and requires more power to operate. Thus the power draw on your PC and on your mouse is increased (+ the additional CPU load due to interrupts), so an engineer's perspective may choose to not implement this since the average practical gain is not really there but saving on power/budget is generally desirable, even if it's just a watt or two. I mean, there actually are reasons why your cheap laptop mouse runs at 125 Hz.
@@theIpatix fantastic points. In fact I suppose it's possible that the cpu overhead could result in a small variance in frame time in cpu bound scenarios. Eliminating the benefit or resulting in a total system latency increase. (Not trying to tell people what is or what isn't just thinking out loud)
Some rig using high polling rate may causing frame drop, so they introduce the option which allows users to change the polling rate when they have this problem.
I've always loved your videos and in depth analysis but this video took things to another level. I appreciate you testing this in CSGO as well. Keep up the good work, one of the best youtube channels out there
Imagine not commenting: ""“Minecraft" "ASMR" "pewdiepie" "music" "Fortnite" "markiplier" “TH-cam is a perfectly balanced game with no exploits.” "Runescape" "World of Warcraft" "Shadowlands" "Dream" "MrBeast" "Warzone" "FaZe Clan" "100 Thieves" "Call of Duty" "Pokemon" "Pokemon cards" "card unboxing" "Charizard" "they don't want you to know" "Flat earth" "round earth" "triangle earth" "the earth is not earth" "what even is earth if not earth omg government is lying to you" "Minecraft" "ASMR" "pewdiepie" "music" "Fortnite" "markiplier" “TH-cam is a perfectly balanced game with no exploits.” "Runescape" "World of Warcraft" "Shadowlands" "Dream" "MrBeast" "Warzone" "FaZe Clan" "100 Thieves" "Call of Duty" "Pokemon" "Halo" "Devil may cry" “TH-cam is a perfectly balanced game with no exploits.” “Cocomelon” “t series” “Minecraft" "ASMR" "pewdiepie" "music" "Fortnite" "markiplier" “TH-cam is a perfectly balanced game with no exploits.” "Reply 4578 if you see this" "Runescape" "World of Warcraft" "Shadowlands" "Dream" "MrBeast" "Warzone" "FaZe Clan" "100 Thieves" "Call of Duty" "Murthox" "Pokemon" "Pokemon cards" "card unboxing" "Charizard" "they don't want you to know" "Flat earth" "round earth" "triangle earth" "the earth is not earth" "what even is earth if not earth omg government is lying to you" "Minecraft" "ASMR" "pewdiepie" "music" "Fortnite" "markiplier" “TH-cam is a perfectly balanced game with no exploits.” "Runescape" "World of Warcraft" "Shadowlands" "Dream" "MrBeast" "Warzone" "FaZe Clan" "100 Thieves" "Call of Duty" "Pokemon" "Halo" "Devil may cry" “TH-cam is a perfectly balanced game with no exploits.” “Cocomelon” “t series
Liked and subbed. This guy knows what he's doing I'm in my last year of electrical and computer engineering and computer science and the testing method holds
the sheer amount of time it took you to gather these insights is insane! Thank you! (I probably should NOT say that I watched the video on 1.25 speed. I am sorry)
A comment due to all your hard work and to help with the algorithm. Interesting to see how the data regarding DPI is different to a lot of advice. Another fantastic video.
Such insanely high quality content. If TH-cam’s algorithm doesn’t push videos like this into user’s recommended sections the TH-cam algorithm is trash.
... ofc it is trash
Thank you!
Sadly I can personally attest that the YT algorithm does not like in-depth technical content like mine. 🙈
@@BattleNonSense sehr schade
@@BattleNonSense "razer viper 8k gone sexual prank the cops came"
This isn't a bath bomb bubble bath challenge. So, obviously algorithm isn't going to push this content. I came here by search lol.
Since some seem to have missed it. At 11:58
I make clear that I do *NOT* recommend to just use the maximum DPI value!
The ideal DPI value depends on many factors like build quality of your mouse, quality of the sensor, firmware, etc.
By simply choosing the maximum DPI value you _might_ make matters worse _if_ your mouse cannot handle that without issues _(jitter, increased delay, etc.)_ .
The point I made there was that very *low* DPI values should be avoided - but don't just go to the other extreme. 😅
Also there will be a followup video where I plan to test more DPI values as well as wired vs. wireless. Not sure about the format of that video or video(s) depending on how many mice I will test.
So a dpi of 800 compare to 1600 will give the same results? Or 1600 dpi is the sweet spot?
Great video. So 800dpi will be same as 1600dpi?
What will you recommended the dpi for Razer Deathadder v2 as afaik it has the latest best sensor may be!
Thanks .
For the people wondering, your 'optimal' dpi depends on the sensor on your mouse, for most 'gaming' sensors the sensor will start smothing/add correction to your imput when it is over 2000 dpi so you should probably stay below that.
@@andreacastelli7066 I cannot say for sure as I have not tested 800DPI
I do plan to do a followup video so that goes on the ToDo list! :)
@@BattleNonSense thanks!
Great work
fancy seeing you here andrew O.O
In the market for another mouse I'm guessing?
yooo andrew whats good
God damn, this channel is such a precious thing. Seriously
its insane how much effort you put in your videos great job man
The most impressive is how you could show detailed tests conditions and conclusions. All perfectly explained in less than 15 mins. Absolute jewel.
oh my god the quality dude you're freaking amazing I'm sure I'm gonna return here like 5 years in the future and I'll say hey I was here 5 years ago, that's my comment over there
You need two more years haha
OBSCENELY high quality deep dive into a very important topic!
I love this channel so much.
Thank you for all of the great work you do, Chris!
as someone who used to play csgo on very low spec laptop, i used to run 250hz on my mouse because it lessens the load on the processor. The processor was so average that when i move the mouse on cs at 1000hz i lost ~10-20 frames (from around 90 to 70-80). By setting it to 250hz, i'm able to maintain 60fps on competitive matches, and no longer lose frames when moving my mouse.
I now use a much better pc, but i probably still wont use 8000hz polling rate even when it's a standard. Just to save on processor power. Also this is an very awesome video, great tests and results!
@BBlenong just buy a better cpu forehead
@@silentplayz6797 ahhh, most old laptops = non replaceable CPU. and often people get them as hand me downs as they often do not have the cash to buy one. so.....
@@madmax2069 i'd say 99.9% of laptops on the market doesn't have the option to swap out cpu or gpu, not just old ones
@@omarcere9383 today thats mainly true, but back in the day (in the core 2 duo days) many laptops had a CPU socket, the only thing you couldn't upgrade was the GPU.
Interesting to know about the fact that low DPI can cause higher latency
If I'm not mistaken it should only really effect the initial movement because it takes more physical movement to trigger a cursor movement.
@@yoyoninjaboy6431 That's kind of what latency means, doesn't it? Like with a Bluetooth mouse, it's an initial movement delay of 8ms over wired mouse, but beyond that the delay doesn't change.
@@filipbunalti yeah I guess I just wanted to state that it would only be relevant if moving a single pixel. Wouldn't it also only be relevant if you reached a breakpoint to process the input on an earlier frame?
@@yoyoninjaboy6431 I'd just like to know if it would make a big difference for me as i have been playing with 800DPI for a long time. Unfortunately Chris didnt show any tests with my exact DPI number so i can only assume a number between 400DPI and 1600DPI
@@yoyoninjaboy6431 Everything you do in terms of mouse motion would be "ahead" by that amount of time i.e. each frame of the game will have more "advanced" mouse position consistently.
This guy and tech Jesus save us everytime :3
So much work went into this testing, respect.
Damnn man 7 weeks huh ? well thank you soo much for making these kinda video that helps us learn more about latency and all the new Tech.
Best viper 8k video I have seen you properly tested it, every other channle that say they test it dont they just say whats on the Razer website and show nice pictures. I have the Viper 8k and I have to move the mouse extreemly fast to hit 7600Hz im very unlickly to hit that speed so the mouse only averages around 4 to 5 k seams like a gimic to me, sure its smooth but so is a good sensor running at 1k Hz.
As an EE, I can appreciate the incredible amount of time/energy put into this.
You don't put 300+ hours into a single video unless you have a true passion.
who else came here from Jackfrags?
PS: Great quality content. I don't regret clicking on this video
absolutly phenomenal content my dude!
you deserve all the subs!
This one video is more useful than a whole year of Linus videos
Patreon donation was worth every penny again.
I don't usually bother with hitting bell button, but it seem TH-cam don't recommend me this insane high quality content even when subscribed smh.
Insane testing, especially if you can automate it. I think you should run a website providing this information for as many mice as possible, I think you would be the only source for this information then.
PLEASE
Well, time to move from 800 to 1600 DPI I guess :D
Thanks for this detailed video, great work!
The problem is that a lot of games have bad sensitivity settings. And when they are bad it's almost always better to have lower dpi because the sensitivity always is too high with high dpi settings.
@@rincx329 Well, maybe that is the case with older games, but any title released after 2015 should do perfectly fine with 1600 DPI and a lower in-game sens.
@@rincx329 maybe we can change the windows sens? mouse 1600dpi windows sens maybe 4 ticks or 5 instead of the default 6. Now the real question then again is would that introduce input lag since i guess the cpu would have to scale the input down so i guess multiply some shit depending on how the scaling is done. Idk how exactly input is processed
@@freesS3 You get less fine-grain control the higher your mouse DPI is, many games give you an arbitrary whole number scale of 1-20 or something like that. For example a game where I play with 400 DPI and '3' in game, and it only has full number steps, if I were to switch to 1600 DPI to maintain the same eDPI I'd want to switch my in game sensitivity to '0.75', which wouldn't be possible.
@@domzaeWoW Well, all the games I play will work perfectly fine for me. I play 30cm/360, so its not the lowest sens. I can easily half my ingame sens in any game I play.
And I wouldn't even know any game which wouldn't allow me to half it either.
If that video leads to most pros changing their dpi, it would be pretty insane. Loving the quality content. Keep it up
It wont because it is flawed. This is an Ad for the razer mouse. While its known that the input delay from higher DPI's is less they are generally speaking less accurate and thats why pro's use lower DPI's. High DPI usually introduces line skipping issues making it overall less accurate.
@@cupcakebunnyprod "line skipping"?
The reason people use lower DPI is that it makes your movement smoother and less jittery. Probably due to higher latency. I personally can track better when my mouse is set to 500Hz even lol. It just makes my movement smoother so I can track easier.
@@chonkusdonkus lines on your screen yes. Higher DPI can skip every other pixel making it less accurate.
@@cupcakebunnyprod there is nothing that would suggest this would happen, unless the sensor is bad and can't track properly
OBSCENELY high quality deep dive into a very important topic! Thank you for all of the great work you do, Chris!
4:24 Because operating at higher poll rates can increase CPU usage and may delay other input chains (like your keyboard), or delay other more important processing chains within your PC.
Having a real 1000hz poll rate is already excessive since your framerate and monitor delay is the main bottleneck to focus on, not your mouse.
i think there's a misunderstanding here. what you explained is a perfectly valid reason to slow down the rate of move events, but b(n)s was asking why one would slow down click events. the key difference is that the sensor can (and will) actually generate 8k move events per second when asked to do so, while the mouse buttons will only generate 1 event per click/release. so unless you perform 8k mouse clicks per second (which is physically impossible) the buttons will never overload the CPU with events and thus don't have to be restricted to a lower polling rate.
so we actually have 2 polling rates now:
1: the actual USB polling rate (determined by the USB port generation)
2: the sensor update rate (determined by the user setting)
except for the G203, in which case the user setting actually lowers #1
@@LastExceed But that's the problem. There's no way to individually set the poll rates for these parameters. You only have one setting within the driver panel (if you're lucky), and that might not even do anything depending on the mouse and its drivers. Nor is there a way for anyone to reasonably verify it, so manufacturers could blatently lie to you.
Which is why I recommend to just set the poll rate to 500 or 1000 and not over, even if the mouse has the capability, since you're trying to get rid of the actual bottlenecks in your system instead of contributing to it.
Since USB devices function on a system-interrupt basis, it really is a trade-off for higher poll rates which can have an impact on other areas of system responsiveness (even framerate in some cases), in exchange for such a minor improvement in mouse responsiveness in realworld scenarios.
Having better frametimes and more consistent FPS is more productive to the overall end-to-end latency when you're already at 500 or 1000hz poll rate.
what if u have 10 cores or higher cpu? does it still matter?
@@TheKillerZmile yes, because input handling is usually single-threaded. Also the actual CPU time requirement of input logic varies vastly from game to game. If its simple and well optimized then it wont be a problem even on mediocre hardware, but in some games even top tier CPUs will suffer
@@TheKillerZmile The amount of cores you have only matters as much as how well modern game engines can utilize them.
We are still stuck in a transitional period where anything beyond 4 or 8 physical/virtual cores is not utilized well and single-threaded performance is still king for gaming.
But even if you could run all IO related tasks like USB processing on a separate core/thread to the game or application, it would still take CPU time to process, and also time to communicate with the game engine.
Incredible video! Thanks for all your hard work!
Insanely high value content. you put so much work into this!
A lot of work in this vid for sure an honor to have subscribed.
I love having the "sharpshooter" button set to an extremely high DPI. It allows for fun things like rapidly spinning or ease of use with controlling certain aspects of games that force your mouse movement slower, like tank turret rotation or helicopters.
Great Vid! I just bought my Razer Viper 8k, I really appreciate you in depth analysis.
For future mouse movement tests, consider strapping the mouse onto a cartesian 3d printer. You should be able to easily automate the movement of the main plate relative to the mouse in 3 dimensions. No need to reset it manually. Probably fast and accurate enough for these tests too.
I considered that.
Following problems:
- how to ensure that the contacts touch
- movement too slow
- expensive
What such a setup (i.e. drawing robot) would be great for is to test sensor accuracy.
@@BattleNonSense What about using a turn table for vinyl records? That would give consistent speed, but the vector would always be the same which is maybe not what you want. Also one must make sure to always put the sensor on the same "track" since speed will vary.
Chris, in your main conclusion you say that a low dpi value (like 400 dpi, which I believe it's the most popular) should be avoided bc your data shows it increases input latency. However, the values you show for the jump from 400 dpi to 1600 dpi are ~ 2 ms for 1000 Hz (again, the most common case) and ~ 4 ms for 8000 Hz, which don't seem all that significant, is this *small reduction (~ 2ms)* worth all the hustle of re-configuring and re-adjusting all the setting of the games and ones cognition? It doesn't seem so to me. Do I missing something?
GREAT WORK! As an engineer you're a great inspiration.
Yes. 2 milliseconds is a lot, you can just alter the sens and have the exact same speed anyways even with changing DPI
love the content!!!
Picoscope FTW! Thanks for your time and effort, it's very much appreciated.
Hey man nice video but why did you not check 800 DPI? I think 400 DPI is known to produce input lag while 800 and anything above should be fine ? im not 100% sure on that but i would love to see an update else i would need to go to 1600 DPI from now on haha
Same 🙄
_"why did you not check 800 DPI?"_
Time - this video took 7 weeks to make.
I will do a followup video though. So more DPI testing is on the todo list.
@@BattleNonSense nice to hear- thanks und dankeschöööön 🤗 / bin sehr interessiert zwischen 800 und 1600 dpi da ich meine razer im Treiber immer auf 800 stehen habe und dann ingame die weiteren Einstellungen vornehme - dank dir habe ich alle g sync Einstellungen + vsync und auch den afterburner auf meinem pc und erstmals spiele ich mit einer gleichmäßigen framtime und der Unterschied ist Gold wert 😘😘😘
@@BattleNonSense Thanks for the great info, can't wait for the follow up vids!
Very impressive work Battle The Razer should 100% add your video to them site !!!
I always felt 1600 dpi was simply better than 400/800 but this pretty much confirms it. More responsive and smoother. The only reason I use 800 is because it's better for the desktop. Now I'm going to try to switch permanently to 1600 for every game that supports it.
WOW great video man, I REALLY enjoyed the technical analysis you did on the mouse!!!!!!!!!!! AMAZING video
8000hz. Highly recommended as an Overwatch tank main like me, especially Reinhard.:) Thx for this stunning work!
lol, there is an option in ow called enhanced mouse precision. If you dont enable that, ow cant even utlize 1000hz.
Your work on this video is insane!
All my questions have been answered.
Thanks a lot.
battle(non)sense: "it is not ideal to use low dpi setting"
every csgo pros : "low dpi goes brrr!!!"
i've always known they're stupid.
not ideal does not mean it will make any difference in a real world environment
@@andreasbarnickel8509 they're stupid
mmm... very salty
It probably makes their aim feel more floaty so they miss less
This was absolutely mind blowing. Thank you so much!
What about 800 vs 1600 dpi for input delay
Exactly my question.
Just use 1600 dpi, no reason to use 800.
@@Rainnnny yeah just change your sensitivity that youve probably been playing on for years. its not that hard bro
i've played on 400dpi for over 7 years and its impossible for me to go up
Yeah he never answered the issue of 800 or 1600 for non 8k hz mice. He assumed the difference was inherent to all mice for the rest of the video.
@@kamiyado1252 just change ingame Sens when you increase dpi , it's not that hard lol
You should receive a Nobel Peace Prize for Outstanding Achievement in Science for this
@Battle(non)sense
It's nice to see you have grown a lot and use lab equipment now.
Thanks to the support of my awesome patrons I am able to afford that gear and not have my channel kill my bank account. 😅
@@BattleNonSense plz test raw input on/off in some games(in cs go and for example apex legends). Some people in reddit say raw input make things worse
And again TY for your work!
Incredible video.
Here is some advise to gamers: don’t fall for the hype of ultra-high mouse polling rates (2000/4000/8000 Hz). It’s just the new marketable number, since no one cares about DPI anymore.
Stick to 1000Hz or even 500Hz to save CPU cycles and mouse battery. It won't affect your gameplay at all. If you really want to reduce your input lag, improve your reaction time with a good sleeping schedule, good diet and exercise. Healthy habits can have a substantial impact in reducing your reaction time.
What a fantastic, thorough investigation! Thank you.
Can I use the pointer speed setting in Windows to get normal cursor speed on the desktop despite using a higher DPI, or are there any issues with that setting?
Windows cursor speed other than the neutral setting introduces mouse acceleration on the desktop and in any software that doesn't use raw input (shouldn't be a problem for most shooters).
This is a good question
@@crownofarnias I have not seen changing the pointer speed in windows introduce mouse acceleration - that is with "enhance pointer precision" *disabled* ofc. :)
Looks like I was misinformed and the reason people recommend 6/11 mouse speed has something to do with skipping counts/pixels, not acceleration. Not sure how valid that is.
Either way, the setting shouldn't affect the aiming in most modern shooters because of raw input. It might affect the parts of the game where you have a cursor though.
So much effort for this video. Amazing!
Commenting for YT algorithm visibility
Great vid
To sum up: 1000hz polling rate + 1600 DPI is the sweet spot (for most people).
i swear this is the most insane video i have ever seen in years
you are a legend dude keep up the good work
Battlenonsense were the mouse pad pushed at the same speed with the two different dpi values? If so the 1600dpi would be moving 4 times as much right? Could that be the reason for the lower latency? Also were you using usb 2.0 or usb 3.+?
Also very curious to know if stuff like the systemresponsive setting in the registry database changes anything for your results? Can't shake the feeling that something like drivers from different mice was the cause of the polling rate not working properly?
I think lower dpi has lower latency because the sensor needs to travel 4 times less distance to send information to the computer that the louse loved but i didn't expect ut could have that much impact
USB 3.0 (aka SuperSpeed) is completely irrelevant for input device latency. The fastest interrupt service interval of Super Speed is 8000 Hz, the same as High Speed USB.
@@Rezler_ i am sure you mean higher dpi and according to my own test it is opposite of what you said.if i move my mouse slowly across the screen the polling rate is very low... somewhere around 15-25hz... if i start moving it faster across the screen it will go up to 1000 hz. It all depended on how faster i was moving the pointer on the screen not the mouse. So if you move a mouse physicly the same distance at the same speed the dpi will play a factor to the polling rate making higher dpi give the result of higher polling rate
@@theIpatix actually i wanted to make sure he was using usb 2.0 as the newer versions have plenty of reported issues with mice
Sir, you are insane and I love it! 🧡💛🧡 Such meticulous test setup is something to behold!
Some people like to lower their polling rate because it interferes with USB power and can cause bugs in some programs. This happened to AntVenom in Minecraft.
This kind of in depth investigation I bet even the mouse manufacturers didn't do this much testing on their mice.
@Battle(non)sense your testing/engineering skills is fucking insane )
I predict that nvidia will use your findings in new feature in RTX 40x0, as they did with your latency work in 30x0 ))
I think they did on the reflex
Your standard of testing + transparency/portrayal of data is still extremely refreshing. Great work :)
Very great video as always, you do so much tests and it's always the right metodology, I think lower dpi has lower latency because the sensor needs to travel less distance to send information to the computer that the louse loved but i didn't expect ut could have that much impact
u think？？Dude already tested it out that low sens has high latency. what are u even saying？
WTF, the way you analyze made me subs to you. This is some next-level review. Keep posting!
ngl i thought your desk was massively deep but it was just optical effect :)
it is quite deep with ~90cm - I built that myself :)
@@BattleNonSense Is it mounted on the wall or you added legs ?
awesome Video and awesome German Quality Engineering with the Idea on how to actually "measure" something that brings Data that is comparable. I would have loved to see the CPU Usage with 8k vs 1k and 500hz because if Razer did no special "Magic" with that 8k processing, the CPU Load should be huge.
I think people would use higher dpi if games wont spin like aimbot when you have lowest sensitivity and 800 dpi.
I also wish every game has a sensitivity slider like csgo where you can enter any value like 0.85 without stepping over it.
Yeah, most games have garbage sensitivity settings. RDR2 on PC comes to mind. It's unplayable with DPI higher than 400.
@@VargVikernes1488 maybe devs use old trackbal mouse
@@VargVikernes1488 good thing aim aint important on RDR
It's so dumb. Even FPS game devs think people want to play 5 cm per 360. If a game has 50 sensitivity notches to choose from, 5 to 50 notches of the ingame sensitivity are uselessly fast. Leaving you with 5 values to choose from between 1-5 so you're forced to change your DPI to fine-tune.
Yea i use 400dpi mostly because every game has a ridiculous sensitivity setting out of the box.
best video i ever seen . GO on and dont stop and you will be 1 mill+ sub for shure!
Did you test the G203’s motion delay? Would be interesting to know how mercury/hero sensor compares to pixart sensors
Sorry but I did not. At one point I had to stop to add more and more tests. 😅
Absolutely brilliant 👍🏻 Thank you.
Thanks to JackFrags for sending me here, so much good stuff !
What video did jackfrags mention this video just curious?
amazing quality content, you definitely deserve more subs
Having 2 little kids increased my input lag tremendously :P
When I think about moving my mouse it takes about 1 day to 2 weeks for it actually moves. But then it is mostly fine for a few minutes.
I'm lucky that mine gets distracted by the pretty lights in the computer usually buys me 30-45 minutes uninterrupted lol
@@moonmessiah9368 Lol :)
Nonono the pretty lights on the keyboard draw the 2 year old in...they have to be pressed/slapped and turned off :)
And when they are off they have to be turned on and off and on and off :P
Before I would get some time during naptime but that is over..no more naps .
Amazing work again. Danke für die perfekten Videos
*Has super accurate 8kz polling rate mouse*
*plays Reinhardt*
Lmfao!
Bought mine and every game works with the 8khz. A lot of people think it doesn't work right and for some it doesn't work. But in the end of 2021 it is worth getting!
Wow, amazing work! But... Wait what? Did Battle(non)sense just proved that 1600 dpi is better than 800 and 400 dpi? Sadly there was no 800 dpi tests which I got used to. Now Im confused and Idk I need to switch to 1600 dpi or 800 dpi is fine? And im talking here about Hero Sensor with 1k Hz (g pro superlight and my g305).
@Battle(non)sense - can u explain to us if 1600 dpi matters only for 8khz? or we should also use higher dpi like at least 1600 for 1khz mouses?
And im talking here about 240hz/1080p (with no more than 240 fps lets say). I think 800 dpi is perfect spot because even if I devined my sensitivity in game I feel my mouse is overshooting.
There will be a followup video where I test more DPI values and (hopefully) wired vs. wireless.
@@BattleNonSense after my subjective tests I think 1600 dpi can matters on 360hz only. For 1080p 240hz 800dpi is still enough. Still I wonder how pros can play on 400dpi and still be good
@@sebash6 1. dpi probably has a negligible difference in their overall performance in the game; 2. they would be even better with good dpi settings.
I'm surprised you don't have a million subs yet!
Lots of people like to use 400 dpi for general windows/application use -- do you expect something like RawAccel (mouse-acceleration programs operating at kernel level) or using something like 4/11 instead of 6/11 in the windows mouse speed settings to effect the input delay when switching to a higher dpi like 1600? If so, to what extent, would it be totally counter productive and eliminate any input lag gains from the switch from 400dpi to 1600dpi? I suppose a qualitative answer is the best to hope for until such a test is performed, but I'm curious to hear your thoughts!
Use 400-800 in Windows and cursor-heavy games and switch to higher DPI for shooters. Changing the settings in Windows itself, unless you're sure the game uses raw input, is asking for trouble.
Keep it up man, This is great work. A video I would love to see is windows settings and the influence it has on the gaming experience. Like windows game bar etc.
So a lower DPI causes an increase in delay? How interesting! This is significant considering the debate between using low DPI/high sens or high DPI/low sens..... I've always been told it doesn't matter which way you go since they "balance out" either way.
Multiple factors could play into this, but a higher DPI will obviously give you better movement resolution over time resulting in less latency. I didn't crunch the numbers, but depending on how fast you actually move the mouse this may or may not have a lot of impact. Further delay may also be added by data processing from the sensor which may depend on the mouse settings. In the end it's always an "it depends".
so much information and testing :O you're doing god's work here, keep it up
What is the impact of high polling when playing at more reasonable framerates (100-200)? This analysis is really impressive but it still doesn't explain what this tech will (or won't) do for most PC gamers. Until 360 hz and up monitors become the norm and GPUs are widely available to put out these kinds of framerates AND games are developed to run at these framerates without glitches and physics issues this all still seems to be creating a solution for a problem that doesn't yet exist.
Marketing mostly but if it costs nothing and is technically better why not.
8000Hz is really meant for those who already maxed out everything else.
But if the mouse you want to buy anyway supports more than 1000Hz then it won't do any harm.
Totally get that and I'm glad they added it to the mouse rather than make it a "premium" feature. Just hope this doesn't turn into polling rate wars between manufacturers like with dpi and sensors in the past. Amazing analysis and quality AS ALWAYS, would just like to see more of what the impact is for more typical fps and any insight content creators might have from razer regarding support for this tech in games going forward.
@@yoyoninjaboy6431 Well, there is a cost in multiple regards: You need a USB High Speed PHY, which is generally more expensive (probably negligible in relation to "gamer" mice prices) and requires more power to operate. Thus the power draw on your PC and on your mouse is increased (+ the additional CPU load due to interrupts), so an engineer's perspective may choose to not implement this since the average practical gain is not really there but saving on power/budget is generally desirable, even if it's just a watt or two. I mean, there actually are reasons why your cheap laptop mouse runs at 125 Hz.
@@theIpatix fantastic points. In fact I suppose it's possible that the cpu overhead could result in a small variance in frame time in cpu bound scenarios. Eliminating the benefit or resulting in a total system latency increase. (Not trying to tell people what is or what isn't just thinking out loud)
You are A CRAZY man...Thank you for yet again another amazing and informative video :)
Lol, RocketJumpNinja could learn from this. This is how you test mice
Holding anyone to BattleNonsense's standard is absurd. Just be grateful for the quality of both.
Different types of testing - this is objective testing, RJN almost exclusively does subjective testing. Both have their place.
@@filipbunalti I am aware, for the timewindow he releases videos he could do that research. Assuming that he has time for that.
Some rig using high polling rate may causing frame drop, so they introduce the option which allows users to change the polling rate when they have this problem.
The majority of CSGO Pros use 400dpi, they should adjust their sensitivities to make their DPI higher and their in game sensitivity lower?
if they want that 2ms of lower latency then the data suggest that they should
and most of them uses 4:3 resolution, i dont know if that has some effect like battlenonsense test.
I think they do it because they used to it.
@@TheKillerZmile Not really. You might get a few extra FPS because you have a lower resulution than 16:9 but the rest stays the same.
@@tunnfisch7548 There could be some scaling latency in play. I'm not saying there is but I see how that could be the case.
Appreciate such effort, Genius work!
gpu scaling vs display? any knowledge, maybe a potential vid
gpu nowadays is faster and neglible on delays
Yes! I too think equal.
Keep up the good work Chris!
Can you make a netcode video about Hunt Showdown? / Kannst du ein Netcode Video über Hunt Showdown machen?
I've always loved your videos and in depth analysis but this video took things to another level. I appreciate you testing this in CSGO as well. Keep up the good work, one of the best youtube channels out there
Imagine not commenting: ""“Minecraft" "ASMR" "pewdiepie" "music" "Fortnite" "markiplier" “TH-cam is a perfectly balanced game with no exploits.” "Runescape" "World of Warcraft" "Shadowlands" "Dream" "MrBeast" "Warzone" "FaZe Clan" "100 Thieves" "Call of Duty" "Pokemon" "Pokemon cards" "card unboxing" "Charizard" "they don't want you to know" "Flat earth" "round earth" "triangle earth" "the earth is not earth" "what even is earth if not earth omg government is lying to you" "Minecraft" "ASMR" "pewdiepie" "music" "Fortnite" "markiplier" “TH-cam is a perfectly balanced game with no exploits.” "Runescape" "World of Warcraft" "Shadowlands" "Dream" "MrBeast" "Warzone" "FaZe Clan" "100 Thieves" "Call of Duty" "Pokemon" "Halo" "Devil may cry" “TH-cam is a perfectly balanced game with no exploits.” “Cocomelon” “t series” “Minecraft" "ASMR" "pewdiepie" "music" "Fortnite" "markiplier" “TH-cam is a perfectly balanced game with no exploits.” "Reply 4578 if you see this" "Runescape" "World of Warcraft" "Shadowlands" "Dream" "MrBeast" "Warzone" "FaZe Clan" "100 Thieves" "Call of Duty" "Murthox" "Pokemon" "Pokemon cards" "card unboxing" "Charizard" "they don't want you to know" "Flat earth" "round earth" "triangle earth" "the earth is not earth" "what even is earth if not earth omg government is lying to you" "Minecraft" "ASMR" "pewdiepie" "music" "Fortnite" "markiplier" “TH-cam is a perfectly balanced game with no exploits.” "Runescape" "World of Warcraft" "Shadowlands" "Dream" "MrBeast" "Warzone" "FaZe Clan" "100 Thieves" "Call of Duty" "Pokemon" "Halo" "Devil may cry" “TH-cam is a perfectly balanced game with no exploits.” “Cocomelon” “t series
Nice content about mice, rarely seen on TH-cam. Thumbs up.
Awesome video chris. Cannot wait for more videos.
Great video Chris. It's clear you put a ton of work into this video.
Liked and subbed. This guy knows what he's doing I'm in my last year of electrical and computer engineering and computer science and the testing method holds
Wow this is exactly what i was looking for since i bought the razor 8k mouse. I subbed n thank you for ur work!
the sheer amount of time it took you to gather these insights is insane! Thank you!
(I probably should NOT say that I watched the video on 1.25 speed. I am sorry)
A comment due to all your hard work and to help with the algorithm. Interesting to see how the data regarding DPI is different to a lot of advice. Another fantastic video.
This is the cheapest way to gain significant millisecond response times on my rig (6700k 1070 gtx). Thank you very much for this superb video
Thank you for all the testing and info. Greatly appreciated.
Amazing review and breakdown !!! You got my sub
How good it is to see your scientific approach in a time when we consume so much unsupported information.