Human Benchmark implemented in C++ DirectX9

Everything about latency. Tips, testing methods, mouse lag, display lag, game engine lag, network lag, whole input lag chain, VSYNC OFF vs VSYNC ON, and more! Input Lag Articles on Blur Busters.
milojr21
Posts: 85
Joined: 23 Jul 2018, 22:46

Re: Human Benchmark implemented in C++ DirectX9

Post by milojr21 » 21 Oct 2020, 21:07

This is really cool! Is it a bug that when you press escape on red, say you already did a few and you press escape when it's red it doesn't show your stats? has to be on black? Also a suggestion maybe have a count indicator to see how many clicks you've done at the top right or something?

User avatar
schizobeyondpills
Posts: 103
Joined: 06 Jun 2020, 04:00

Re: Human Benchmark implemented in C++ DirectX9

Post by schizobeyondpills » 25 Oct 2020, 14:56

Meowchan wrote:
14 Oct 2020, 02:12
Have you tried both left mouse button and the 'b' and 'w' buttons on the keyboard?

As for save function I don't know how much demand there is for the tool. I agree obtaining the results is cumbersome with the screenshot and the alt tab, but not sure if the extra tweaking can be justified at this point.
why not go Direct X 11 and implement all of my suggestions, also a proper reaction time test would have MIN random delay be atleast top 0.1% focus time, so that people cant cheat. https://docs.microsoft.com/en-us/window ... amelatency

User avatar
xeos
Posts: 43
Joined: 12 Jul 2018, 14:54
Contact:

Re: Human Benchmark implemented in C++ DirectX9

Post by xeos » 25 Oct 2020, 15:23

schizobeyondpills wrote:
25 Oct 2020, 14:56

why not go Direct X 11 and implement all of my suggestions, also a proper reaction time test would have MIN random delay be atleast top 0.1% focus time, so that people cant cheat. https://docs.microsoft.com/en-us/window ... amelatency
I am also concerned there might be cheating, but I'm not sure how this helps?
Measure display input lag the cheap way or the best way (IMHO, but I'm biased).

User avatar
schizobeyondpills
Posts: 103
Joined: 06 Jun 2020, 04:00

Re: Human Benchmark implemented in C++ DirectX9

Post by schizobeyondpills » 25 Oct 2020, 15:30

xeos wrote:
25 Oct 2020, 15:23
schizobeyondpills wrote:
25 Oct 2020, 14:56

why not go Direct X 11 and implement all of my suggestions, also a proper reaction time test would have MIN random delay be atleast top 0.1% focus time, so that people cant cheat. https://docs.microsoft.com/en-us/window ... amelatency
I am also concerned there might be cheating, but I'm not sure how this helps?
in WoW, pre-potting is a term where you prepare yourself and use stat boosting potion before fight starts, same is applied here, biggest cheating u can do is pre-fire to 99% of activation on your nerves/hand/arm/finger/mouse button/mental response, but you can only keep your focus in that state for so long, after certain long period you will lose it because its a limit of human mind that it cannot remain in a highly focused low reaction state for long, lower min random interval means u can cheat by entering this high focus instantaneous pre-calculated state.

User avatar
Chief Blur Buster
Site Admin
Posts: 11647
Joined: 05 Dec 2013, 15:44
Location: Toronto / Hamilton, Ontario, Canada
Contact:

Re: Human Benchmark implemented in C++ DirectX9

Post by Chief Blur Buster » 25 Oct 2020, 16:48

schizobeyondpills wrote:
25 Oct 2020, 15:30
in WoW, pre-potting is a term where you prepare yourself and use stat boosting potion before fight starts, same is applied here, biggest cheating u can do is pre-fire to 99% of activation on your nerves/hand/arm/finger/mouse button/mental response, but you can only keep your focus in that state for so long, after certain long period you will lose it because its a limit of human mind that it cannot remain in a highly focused low reaction state for long, lower min random interval means u can cheat by entering this high focus instantaneous pre-calculated state.
AFAIK, the C++ Human Benchmark is already doing randomization (1 second jitter).

Please correct me if I am wrong.
Meowchan wrote:
17 Aug 2020, 18:08
Benchmark.DX9.Human.exe - Uses mouse buttons (and Esc to quit). First click turns screen red, after a random 3-4 seconds delay turns screen green similar to the popular Human Benchmark website. It keeps stats of min/max/avg delay as well as how many early clicks you had. Needless to say it is much faster than the web version.
Head of Blur Busters - BlurBusters.com | TestUFO.com | Follow @BlurBusters on Twitter

Image
Forum Rules wrote:  1. Rule #1: Be Nice. This is published forum rule #1. Even To Newbies & People You Disagree With!
  2. Please report rule violations If you see a post that violates forum rules, then report the post.
  3. ALWAYS respect indie testers here. See how indies are bootstrapping Blur Busters research!

User avatar
schizobeyondpills
Posts: 103
Joined: 06 Jun 2020, 04:00

Re: Human Benchmark implemented in C++ DirectX9

Post by schizobeyondpills » 25 Oct 2020, 16:54

Chief Blur Buster wrote:
25 Oct 2020, 16:48
schizobeyondpills wrote:
25 Oct 2020, 15:30
in WoW, pre-potting is a term where you prepare yourself and use stat boosting potion before fight starts, same is applied here, biggest cheating u can do is pre-fire to 99% of activation on your nerves/hand/arm/finger/mouse button/mental response, but you can only keep your focus in that state for so long, after certain long period you will lose it because its a limit of human mind that it cannot remain in a highly focused low reaction state for long, lower min random interval means u can cheat by entering this high focus instantaneous pre-calculated state.
AFAIK, the C++ Human Benchmark is already doing randomization (1 second jitter).

Please correct me if I am wrong.
Meowchan wrote:
17 Aug 2020, 18:08
Benchmark.DX9.Human.exe - Uses mouse buttons (and Esc to quit). First click turns screen red, after a random 3-4 seconds delay turns screen green similar to the popular Human Benchmark website. It keeps stats of min/max/avg delay as well as how many early clicks you had. Needless to say it is much faster than the web version.
im saying its not enough because people can stay in this pre-reacted state for atleast 10-15seconds, so that should be the minimum delay. its different to measure human reaction time on unexpected event and other thing to measure it when they know they are being measured and want to do best at it (so then their brain enters this pre-reacted up to 99% state), however you cant stay in it for long (thing about it like a sniper holding its breath and focus)

User avatar
Chief Blur Buster
Site Admin
Posts: 11647
Joined: 05 Dec 2013, 15:44
Location: Toronto / Hamilton, Ontario, Canada
Contact:

Re: Human Benchmark implemented in C++ DirectX9

Post by Chief Blur Buster » 25 Oct 2020, 17:46

schizobeyondpills wrote:
25 Oct 2020, 16:54
im saying its not enough because people can stay in this pre-reacted state for atleast 10-15seconds, so that should be the minimum delay. its different to measure human reaction time on unexpected event and other thing to measure it when they know they are being measured and want to do best at it (so then their brain enters this pre-reacted up to 99% state), however you cant stay in it for long (thing about it like a sniper holding its breath and focus)
True. Then it needs to be configurable, with a watermark (so settings shows in screenshots/photos).

I prefer 1 second because:

(A) Short attention spans of the YouTuber generation.
e.g. Make it boring and you won't have suffiicent firehose of delicious data from lots of population;

(B) Many esports games have frequent 1-second pre-prime anyway,
e.g. You hear approaching footsteps, so you're already primed to shoot when enemy appears around a corner. This is pretty common

The 1-second randomization prevents the predictive guessing sufficiently if you're honest about not cherrypicking screenshots. Reliable pre-prime successes is not the same thing as guessing (chance). And if people really do want to show off non-cheat attempts, post consecutive screenshots (watermark could show attempt #, saved across program executions and installs too in an encrypted registry key). Or whatever.

Yes, anybody doing PhotoShopping or software-reinstalling might still be a problem, but it prevent low-effort casual cheating (like keyholes or thumbscanners; people can still copy physical keys or fingerprints, even from photographs of keys or photographs of fingerprints, but it takes effort to do so, much like it takes effort to hunt down and remove an encrypted registry key or effort flawlessly photoshopping a good textured-digits watermark. Heck, even use steganography techniques to guarantee detection of photoshopping.
Anybody that wants to cheat much is probably already caught cheating in other things anyway...). Low lying apples that stays within the parameters of the current situation of mainstreamer practicalities;

This might mean two different human-benchmarking standards, but both 1-second and 15-second randomization have merit even if some people disagree;

There is merit in both the honest pre-primed benchmark and the honest non-pre-primed benchmark.
Head of Blur Busters - BlurBusters.com | TestUFO.com | Follow @BlurBusters on Twitter

Image
Forum Rules wrote:  1. Rule #1: Be Nice. This is published forum rule #1. Even To Newbies & People You Disagree With!
  2. Please report rule violations If you see a post that violates forum rules, then report the post.
  3. ALWAYS respect indie testers here. See how indies are bootstrapping Blur Busters research!

User avatar
xeos
Posts: 43
Joined: 12 Jul 2018, 14:54
Contact:

Re: Human Benchmark implemented in C++ DirectX9

Post by xeos » 25 Oct 2020, 20:08

Chief Blur Buster wrote:
25 Oct 2020, 17:46

This might mean two different human-benchmarking standards, but both 1-second and 15-second randomization have merit even if some people disagree;

There is merit in both the honest pre-primed benchmark and the honest non-pre-primed benchmark.
I'm sure the 15-second delay would have shocking long delays in response, which would be kind of interesting. But since we are talking about gaming here, how often do you play a game with no input for 15 seconds straight? Or with no visual change for 15 seconds straight? A 1 second delay is much more informative from a gaming perspective unless we are talking chess ;-)
Measure display input lag the cheap way or the best way (IMHO, but I'm biased).

Meowchan
Posts: 40
Joined: 17 Jun 2020, 02:06

Re: Human Benchmark implemented in C++ DirectX9

Post by Meowchan » 25 Oct 2020, 20:41

The high intensity reaction time is what I was going for and what I am curious about. If the time window was larger, say 5, 15, 60 seconds, who's to say there are no peaks and valleys of your attention span and you just so happened to luck out and be extra focused on the narrow window when the screen changed? It would probably be longer reaction time, but will it be consistently longer by a certain amount? The way I see it having a small window minimizes variables and in no way simulates a sniper holding a spot scenario in a game. It also hopefully minimizes the human factor, which makes the machine factor comparatively the largest it's going to get on this sort of test. If you happen to have an optimized and a non-optimized machine available for testing the difference can be quite telling.

User avatar
xeos
Posts: 43
Joined: 12 Jul 2018, 14:54
Contact:

Re: Human Benchmark implemented in C++ DirectX9

Post by xeos » 31 Oct 2020, 13:50

Meowchan wrote:
25 Oct 2020, 20:41
The high intensity reaction time is what I was going for and what I am curious about. If the time window was larger, say 5, 15, 60 seconds, who's to say there are no peaks and valleys of your attention span and you just so happened to luck out and be extra focused on the narrow window when the screen changed? It would probably be longer reaction time, but will it be consistently longer by a certain amount? The way I see it having a small window minimizes variables and in no way simulates a sniper holding a spot scenario in a game. It also hopefully minimizes the human factor, which makes the machine factor comparatively the largest it's going to get on this sort of test. If you happen to have an optimized and a non-optimized machine available for testing the difference can be quite telling.
what would be ideal is to occasionally throw in a much longer delay, say 4 sec in 1 out of every 5 trials (randomly distributed, of course). Then you could see what happens when the expected cadence of response is violated. I suspect that small change would hurt response times both on the longer delay trials and the short one too. Knowing the narrow window in which you are going to respond allows motor system preparation and can really reduce the delay in a simple button pushing experiment.

This is a rich topic. I wouldn't be too quick (ha, ha) to dismiss the scientific literature on reaction time. There are a lot of issues to consider.
Measure display input lag the cheap way or the best way (IMHO, but I'm biased).

Post Reply