[OLD] Non-CPU/GPU Bound Input Lag Tests

[jorimt]

Edit: Looks like I'm actually dealing with a child here, probably from one of those gaming tweak communities. Why do we actually keep such people on the forum? They add absolutely nothing.

He's been moderated before. I'll leave that to the Chief to decide, however.

With the tone of your recent posts (in this and other threads) lately, don't be surprised if further action is taken. You don't have to be ceaselessly insulting to get your points across.

We get it, you don't think high speed camera tests are adequate for measuring sub ms differences, and depending on the experience and knowledge of the individual tester, and the refresh rate/high speed camera FPS in question (240Hz monitor + 240 FPS high speed camera is virtually pointless, as is 1000Hz monitor + 1000 FPS camera, for instance; 1000 FPS capture, phone or otherwise, will certainly be less and less effective the higher Hz consumer displays become, and we're getting there), your sentiment can be correct. The issue here is, it "can" be, but it isn't always.

[Eonds]

Edit: Looks like I'm actually dealing with a child here, probably from one of those gaming tweak communities. Why do we actually keep such people on the forum? They add absolutely nothing.

He's been moderated before. I'll leave that to the Chief to decide, however.

With the tone of your recent posts (in this and other threads) lately, don't be surprised if further action is taken. You don't have to be ceaselessly insulting to get your points across.

We get it, you don't think high speed camera tests are adequate for measuring sub ms differences, and depending on the experience and knowledge of the individual tester, and the refresh rate/high speed camera FPS in question (240Hz monitor + 240 FPS high speed camera is virtually pointless, as is 1000Hz monitor + 1000 FPS camera, for instance; 1000 FPS capture, phone or otherwise, will certainly be less and less effective the higher Hz consumer displays become, and we're getting there), your sentiment can be correct. The issue here is, it "can" be, but it isn't always.

Thanks for explaining. My initial responses to everyone (recently) was respectful until they started insulting me. Accusing me of racism & saying I'm stupid. So please don't overlook the simple fact that they're the ones starting with the inflammatory language. I simply told this guy to watch his mouth because I haven't even said anything towards him. I'm saying your testing methodology is inaccurate. The fact that you're offended by that and resorted to insulting me & then crying. If you really cared about your testing you wouldn't of resorting to insulting me about calling out your testing. Keep it scientific please.

[jorimt]

Thanks for explaining. My initial responses to everyone (recently) was respectful until they started insulting me. Accusing me of racism & saying I'm stupid. So please don't overlook the simple fact that they're the ones starting with the inflammatory language. I simply told this guy to watch his mouth because I haven't even said anything towards him.

I'm not overlooking the (relative) lack of insulting language in your initial replies lately, but you're still effectively telling people their contributions are pointless, and in essence, they don't know what they're talking about.

Further, since you have no notable experience in click-to-photon testing, you have no alternatives to offer beyond the vague "multi-billion dollar companies have better equipment than you; give up, we can never know anything with your limited data because I say so."

This will undoubtedly garner hostile responses, at which point you should agree to disagree, and leave them to the moderators. Any other action will ultimately result in a fruitless argument that can escalate to locked threads and bans.

If they don't like what you said the first time, repeating it subsequent times in continually harsher language will do the opposite of any convincing, which I assume is your (and usually most anyone's) ultimate aim.

[Eonds]

Thanks for explaining. My initial responses to everyone (recently) was respectful until they started insulting me. Accusing me of racism & saying I'm stupid. So please don't overlook the simple fact that they're the ones starting with the inflammatory language. I simply told this guy to watch his mouth because I haven't even said anything towards him.

I'm not overlooking the (relative) lack of insulting language in your initial replies lately, but you're still effectively telling people their contributions are pointless, and in essence, they don't know what they're talking about.

Further, since you have no notable experience in click-to-photon testing, you have no alternatives to offer beyond the vague "multi-billion dollar companies have better equipment than you; give up, we can never know anything with your limited data because I say so."

This will undoubtedly garner hostile responses, at which point you should agree to disagree, and leave them to the moderators. Any other action will ultimately result in a fruitless argument that can escalate to locked threads and bans.

If they don't like what you said the first time, repeating it subsequent times in continually harsher language will do the opposite of any convincing, which I assume is your (and usually most anyone's) ultimate aim.

Well unfortunately it's apart of science. You have to be told when you're doing something wrong. I'm not trying to tear anyone down but it's literally not worth the time and effort for anyone who doesn't have plenty of money and high level skills in multiple areas. There's no cheap alternative & to say i'd know how to fully rig up a perfect testing scenario would be false. I have an idea but that requires not only specially designed equipment but special environments. We can measure multi millisecond differences relatively accurately with the current cheaper methods. If the person gets hostile because of this they need to grow up honestly. I try to sometimes show people their flawed logic in a non hostile response to further the conversation into something constructive. Instead I get insulted which shows how non-serious the poster is about the topic at hand. These specific things have been tested already & literally developed to lower latency. So unfortunately that's the actual scenario and I don't mean to offend anyone but if I did I don't care because that's your personal ego problem you should deal with outside of the forum. I'm being vague regarding a methodology because it would take multi-thousand dollar scopes & probing, perfect CUSTOM hardware, software, firmware, likely would need multiple people to assist you since you probably don't have professional level knowledge in 20 different topics. You would need a perfectly sterile and almost EMI free lab esk area. It's so complex that I can't even think straight. There's so many things you'd need to do to properly complete such a task and it's being done by big companies for a reason. If gerald gizzard bob the 3rd could run a good o'l 20$ method and get accurate results don't you think they'd just do that and not throw billion(s) of dollars into this. Yes it sucks that we have to rely on them for truly accurate results (assuming it's not manipulated). The best we can do is use common sense, documentation, ask qualified people on the topic, draw obvious conclusions, be objective & leave the ego behind. (that's a general statement to anyone reading it). My intent is not to be the teacher that yells at the kid who's clearly trying his best to complete a test. I'm simply saying, Hey that's wrong, here's why & what you can/can't do.

[jorimt]

Unfortunately, any established science is a consensus of its time, and consensus can also be objectively wrong. Only truth is absolute, and even what people consider "truth" can be subjective, so good luck getting everyone you come across to concede to 100% of your input.

In respect to this thread in particular, if more than 1ms difference isn't consistently showing in the accumulative average latency readings between scenarios, it's safe to say that if those settings do make any impact, they're probably below 1ms on average.

If you're getting caught on the fact that you feel he's claiming he knows without a doubt those settings do nothing for latency, even at the sub ms-level, then again, make your point, and agree to disagree, because neither of you can prove it one way or the other.

Same goes for me; I can't, which is why I've only ever limited my testing to render queue and sync latency. Removing the extra ~17ms on my 240Hz setup by avoiding render queue and sync latency is personally enough to sustain my gaming hobby. I'll let others squabble over the remaining sub-ms gains that I find too often have a ridiculous benefit/cost ratio.

[MT_]

our system is semi-stock. You're still on windows with a high standard deviation that cant measure sub ms differences concretely. You're not doubt competent to some degree so I wont be too harsh. The equipment which I'm talking about is hundreds of thousands of dollars which mostly companies like AMD & Intel & Nvidia use. I don't see the point of this post other than to look cool. It's true that most people have terrible systems. I think if you're not an actual expert on the topic you shouldn't be posting results that are inconclusive & inaccurate. Every BIOS (basically) has SMI's. They do cause latency spikes. This is !ONE! out of THOUSANDS of factors which you cannot change UNLESS you're HIGHLY knowledgeable. You'll only see what they want you to see. They don't care about you or anyone. A 1000 FPS camera is not good enough either. Do you think a company that has spent billions of dollars onto measuring latency would use a 1000 FPS camera to measure system latency ON WINDOWS? "Gaming Tweak Communities" says the drone shit posting on blurbusters with useless information and inaccurate results. Please don't talk shit to me, I'm trying to be constructive. In reality we both know why you made this post & it's cringe. Also gamemode itself has been changed over time through different windows version. You have hundreds of power saving features & hundreds of other useless clock gating/power gating things bogging down your GPU + thousands of other factors. It's hilarious because nothing of which I said was false & you still manage to turn into a twitter tweaker because I pointed out your false measurements. This is the problem with Science usually, EGO gets in the way. I don't care how you feel, I care about actual measurements. Leave your shitty attitude else where. Talk about adding nothing, look at your post dude....

I'll enlighten you, I'm not here to seek sub millisecond input latency differences so maybe this wasn't clear to you. But I'm under the impression that you do not entirely grasp the idea behind my tests. I was trying to see if there was any noticable or measurable difference in the normal millisecond range. (Considering the fact that I mentioned i was using a 960 fps camera, this should've been obvious to you, yet you didn't even grasp that.)

As for your claim that 'MSI Mode' option is 'designed' to reduce latency, and I assume you refer to my tests that include 'MSI Mode', obviously that test is referring to the registry modification of a certain PCI device and its modified operation. If it was solely created to lower latency, why do you think certain devices go haywire when forcing this registry addition causing blue screens or distortion of sound (Most soundcards i.e.). In fact, if a device is not using MSI by default, decided by the driver, the whole registry entry is absent unless you either add it yourself or use one of these tools like "MSI Util". Or do you also accept the fact that MSI mode was 'designed' to make systems unbootable or introduce other quirks?

If you are talking about Message signaled interrupts in general; It was most likely initially invented to get rid of the old physical interrupt lines and IRQ sharing. (Yes indirectly there will be latency reductions).

But the 'MSI Mode' you are referring to is a part of Windows's driver system to allow manufacturers or drivers to dictate whether they want to use native MSI mode or emulated legacy line-based interrupt method. Sufficed to say both go over the same PCIe bus as there are no physical IRQ lines anymore. And if your system is properly configured, no device will share interrupts. If it does (In emulated line-based interrupt mode), then yes in that case there might be a noticable improvement to be seen, but even this depends entirely on what other device is sharing the same IRQ in relation to execution times, PCIe LTR, etc.

You have zero knowledge about anything, stop wasting our time. Blocked.

[Jaketown]

Please get this guy banned from here.

+1, please ban Eonds, he trolls and insulting others and then he cries that it was him who was insulted.

[Eonds]

our system is semi-stock. You're still on windows with a high standard deviation that cant measure sub ms differences concretely. You're not doubt competent to some degree so I wont be too harsh. The equipment which I'm talking about is hundreds of thousands of dollars which mostly companies like AMD & Intel & Nvidia use. I don't see the point of this post other than to look cool. It's true that most people have terrible systems. I think if you're not an actual expert on the topic you shouldn't be posting results that are inconclusive & inaccurate. Every BIOS (basically) has SMI's. They do cause latency spikes. This is !ONE! out of THOUSANDS of factors which you cannot change UNLESS you're HIGHLY knowledgeable. You'll only see what they want you to see. They don't care about you or anyone. A 1000 FPS camera is not good enough either. Do you think a company that has spent billions of dollars onto measuring latency would use a 1000 FPS camera to measure system latency ON WINDOWS? "Gaming Tweak Communities" says the drone shit posting on blurbusters with useless information and inaccurate results. Please don't talk shit to me, I'm trying to be constructive. In reality we both know why you made this post & it's cringe. Also gamemode itself has been changed over time through different windows version. You have hundreds of power saving features & hundreds of other useless clock gating/power gating things bogging down your GPU + thousands of other factors. It's hilarious because nothing of which I said was false & you still manage to turn into a twitter tweaker because I pointed out your false measurements. This is the problem with Science usually, EGO gets in the way. I don't care how you feel, I care about actual measurements. Leave your shitty attitude else where. Talk about adding nothing, look at your post dude....

I'll enlighten you, I'm not here to seek sub millisecond input latency differences so maybe this wasn't clear to you. But I'm under the impression that you do not entirely grasp the idea behind my tests. I was trying to see if there was any noticable or measurable difference in the normal millisecond range. (Considering the fact that I mentioned i was using a 960 fps camera, this should've been obvious to you, yet you didn't even grasp that.)

As for your claim that 'MSI Mode' option is 'designed' to reduce latency, and I assume you refer to my tests that include 'MSI Mode', obviously that test is referring to the registry modification of a certain PCI device and its modified operation. If it was solely created to lower latency, why do you think certain devices go haywire when forcing this registry addition causing blue screens or distortion of sound (Most soundcards i.e.). In fact, if a device is not using MSI by default, decided by the driver, the whole registry entry is absent unless you either add it yourself or use one of these tools like "MSI Util". Or do you also accept the fact that MSI mode was 'designed' to make systems unbootable or introduce other quirks?

If you are talking about Message signaled interrupts in general; It was most likely initially invented to get rid of the old physical interrupt lines and IRQ sharing. (Yes indirectly there will be latency reductions).

But the 'MSI Mode' you are referring to is a part of Windows's driver system to allow manufacturers or drivers to dictate whether they want to use native MSI mode or emulated legacy line-based interrupt method. Sufficed to say both go over the same PCIe bus as there are no physical IRQ lines anymore. And if your system is properly configured, no device will share interrupts. If it does (In emulated line-based interrupt mode), then yes in that case there might be a noticable improvement to be seen, but even this depends entirely on what other device is sharing the same IRQ in relation to execution times, PCIe LTR, etc.

You have zero knowledge about anything, stop wasting our time. Blocked.

Everything you just said is already well known. This is common sense. Stop trying to swing your metaphorical balls around. You aren't above people because you shit post on the forums. 99% of what you just replied to me with is something which i'm familiar with already. MSI was literally introduced for the exact reason I said. Stop trying to discredit what I'm saying it's so weird. With that being said you can't refute anything I said. Your tests were unfortunately pointless & didn't prove or disprove anything useful. I'm glad you still resort to throwing insults and say nothing of use. Try to provide utility or some accurate evidence with your next post please. I didn't need a recap about the PCIE which makes me think you're just trying to pretend to be an authority on the topic. You're trying to impress people but you failed & coming at me was a mistake on your end. Again, leave your ego elsewhere when speaking on here.

[jorimt]

I foresee a thread lock in the near future guys...

[Thatweirdinputlag]

I foresee a thread lock in the near future guys...

Why? The guy @MT_ is voluntarily sharing some results of his own little experiment. I for one was interested! Even though @Eonds observation wasn't completely wrong but it was conveyed with a condescending tone coupled with a hint of arrogance. If he actually writes like that in every reply under his infamous label of "Constructive Criticism" then I can see why he'd be called names. So, please keep this open if possible.

@MT_ When you said "you've bound Background Processes cores 1&2, how did you do that? Manually?