Jump to content



Photo

  • Please log in to reply
155 replies to this topic

#91 WonG.

WonG.

    Neowinian

  • Joined: 07-February 14

Posted 09 February 2014 - 19:32

thanks a lot for your efforts. just for better understanding, im talking about the pictures in post #77. there is a link, but i'm not able to find this software anywhere!




#92 +snaphat (Myles Landwehr)

snaphat (Myles Landwehr)

    Electrical & Computer Engineer

  • Tech Issues Solved: 29
  • Joined: 23-August 05
  • OS: Win/Lin/Bsd/Osx
  • Phone: dumb phone

Posted 09 February 2014 - 19:49

thanks a lot for your efforts. just for better understanding, im talking about the pictures in post #77. there is a link, but i'm not able to find this software anywhere!

Oh, that particular piece of software is private. Comments on the website of that thread asked for it, and the author declined to give it for some reason. As such, the results should be taken with a grain of salt since they are not verifiable.

 

You may be able to find another piece of software that gives the resolution for the other timers, but you can test HPET using wintesttimer.



#93 Eric

Eric

    Neowinian Senior

  • Tech Issues Solved: 13
  • Joined: 02-August 06
  • Location: Greenville, SC

Posted 09 February 2014 - 20:27

I thought that was WinTestTimer and CPU-Z in post 77... oops. It looks like it.

#94 Eric

Eric

    Neowinian Senior

  • Tech Issues Solved: 13
  • Joined: 02-August 06
  • Location: Greenville, SC

Posted 09 February 2014 - 21:00

If your post is removed for any reason then do not repost it. PM a staff member if you want to discuss it.

#95 The_Decryptor

The_Decryptor

    STEAL THE DECLARATION OF INDEPENDENCE

  • Tech Issues Solved: 5
  • Joined: 28-September 02
  • Location: Sol System
  • OS: iSymbian 9.2 SP24.8 Mars Bar

Posted 10 February 2014 - 03:28

Don't discount your Bioshock results. Bioshock is real world benchmark and is exactly what you would want test here to see if HPET results in tangible real-world performance gains/losses. It's the difference between micro-benchmarking and benchmarking actual applications. See: http://en.wikipedia....mark_(computing) and see post #73 in this thread where I discussed the numerous issues with how the original benchmarking was done in this thread.
 
In this particular case with your DPC benchmarks, we are talking about DPC latencies with a few millionths of a second difference on average. These results are so tiny that it would be impossible to perceive a difference for real-time audio/video. Now if there had been large spikes that would be a different story.
 
Also, one last things to note, the LPC Bus is separate from clock generation (It's not the clock source).
...


Yeah, this is one of those things where benchmarking doesn't help much, since any improvements won't be visible. The default timer resolution on Windows is only 16ms (You can lower it, but Microsoft recommends against it for a bunch of reasons), so seeing an improvement of a few microseconds on one benchmark isn't going to translate into a real world difference.

And the benchmark that is useful (The Bioshock one) is noisy enough to not actually show any improvements, and that's just a fact of how computers work (The game isn't the only thing running, each run of the game will actually be different due to multiprocessing, etc.). If we saw a large improvement over multiple runs of the benchmark then we could see that the tweak had some actual effect.

#96 WonG.

WonG.

    Neowinian

  • Joined: 07-February 14

Posted 10 February 2014 - 10:00

These are my results:

 

 

cbv9.jpg

wnfz.jpg

 

My Hardware: 3770K std clock on a GA-Z77-DS3H (TimerResolution.exe running in background) (HPET on, OS and EFI)

 

I saw, for example on Cartels pictures (#60), people with way lower latencies. How to reach them? With HPET off I'm still not able to reach these low latencies... :(



#97 Wakers

Wakers

    Neowinian Senior

  • Joined: 30-July 07

Posted 10 February 2014 - 15:11

In what scenario does it possibly matter?

 

There are no actual use benefits from a lower latency here.



#98 WonG.

WonG.

    Neowinian

  • Joined: 07-February 14

Posted 10 February 2014 - 15:38

In what scenario does it possibly matter?

 

There are no actual use benefits from a lower latency here.

 

 

Hi there,

 

in Germany it's very rude to answer a question with a counter question. But I'd like some help, so I'll try to explain. Before I experimented with the HPET Option activated, I had a lower latency. Many things ran smoother and more fluent. Now with HPET on, some other things work better. So why shouldn't I try to combine these options? Obviously this should be the best possible system setting for me.

 

Kind regards,

WonG.



#99 +snaphat (Myles Landwehr)

snaphat (Myles Landwehr)

    Electrical & Computer Engineer

  • Tech Issues Solved: 29
  • Joined: 23-August 05
  • OS: Win/Lin/Bsd/Osx
  • Phone: dumb phone

Posted 10 February 2014 - 19:16

Hi there,

 

in Germany it's very rude to answer a question with a counter question. But I'd like some help, so I'll try to explain. Before I experimented with the HPET Option activated, I had a lower latency. Many things ran smoother and more fluent. Now with HPET on, some other things work better. So why shouldn't I try to combine these options? Obviously this should be the best possible system setting for me.

 

Kind regards,

WonG.

My former co-worker from Germany use to do this :laugh:. That's a bit of a digression but there are many cases where asking a counter question is perfectly valid. In this case, Wakers has a valid point and question. The reason being that you are seeing latencies in the 500 millionths of a second range. The resolution is so tiny that you should be able to handled deferred real time video and audio without delays (read as: smooth playback).

 

That being said to answer your question, you probably aren't going to be able to achieve lower latencies than you are currently seeing. The only thing else you can do is play with turning off device drivers. For example, turning off wireless will lower my DPC latencies. One other things you can do that I don't personally recommend is disable dynamic scaling of cores (in the power options set the minimum processor state to 100%).

 

Let me be clear about what happens when you have the platformclock setting as false. It's not that it disables HPET, what really happens is that it Windows uses a combination of timers for timing (TSC + HPET). Why? Because using the TSC is cheaper to sample than HPET so it is better suited for scheduling operations. However, at the same time HPET is useful for scheduling one-shot interrupts for sleeping or idling systems (it saves power). The point, I'm making is that different timers have advantages and disadvantages in certain scenarios and this is not, "HPET is the best timer on the system" as the information you'll find around the web would seem to indicate.

 

As for how that relates back to DPC latencies: I noticed a consistent difference of only 50-100us minimum on my laptop between forcing / not forcing HPET. It was far too low make any perceivable difference. To put that in context, in testing the difference is ~1-2% of the maximum latencies I see on my laptop (wifi causes spiking of 1000-3000us at times).

 

To be perfectly honest, I'm not even convinced that the timer they are using to measure DPC has consistent reliable results between when HPET is forced and not forced. Why? The Windows API doesn't give you a consistent clock source. It depends on the system configuration which we are changing in this scenario. See section 2.1.4 here: http://www.windowsti...com/description -- so how can I be sure the differences aren't accounted for by changes in the timer function behavior itself?



#100 WonG.

WonG.

    Neowinian

  • Joined: 07-February 14

Posted 10 February 2014 - 22:54

But I already saw way lower latencies on other screenshot, for example in post #60. Without HPET activated my latency was about 60-90µs. And believe me, the difference is huge (especially in video games). I saw people with latencies about 20-50 with HPET activated. A stable latency (around 100µs) should work for me. Otherwise I'll have to deactivate HPET again, because the disadvantages outbalance the advantages. That's not what I'm aiming for, I don't want to miss the advantages I already acquired :(



#101 +snaphat (Myles Landwehr)

snaphat (Myles Landwehr)

    Electrical & Computer Engineer

  • Tech Issues Solved: 29
  • Joined: 23-August 05
  • OS: Win/Lin/Bsd/Osx
  • Phone: dumb phone

Posted 10 February 2014 - 23:50

But I already saw way lower latencies on other screenshot, for example in post #60. Without HPET activated my latency was about 60-90µs. And believe me, the difference is huge (especially in video games). I saw people with latencies about 20-50 with HPET activated. A stable latency (around 100µs) should work for me. Otherwise I'll have to deactivate HPET again, because the disadvantages outbalance the advantages. That's not what I'm aiming for, I don't want to miss the advantages I already acquired :(

Sorry, I'm not going to believe you without evidence because it wouldn't be very scientific of me. If you are seeing improvements in some tasks with either HPET enabled or disabled, then please provide benchmark evidence showing the FPS or other application improvements. Every claim thus far has been said anecdotally without evidence, but we do have some evidence showing that HPET doesn't result in tangible differences. Until someone provides evidence otherwise, I'm of the presumption that claims of better performance in either case are simply the result of placebo effect and logically this checks out because as I said before, these latency ranges are so low that the processing speed of a deferred call would still provide smooth playback of real-time audio or video. I mean at the end of the day, a typical GPU target framerate is 1 frame every 16666.7 microseconds (60 hertz). The deferred call latency you are seeing is orders of magnitude lower than that. Or to put it another way, a deferred GPU call would be handled 33 times that rate at the minimum (Or if we were targeting 30 hertz, it is 66 times faster than that rate). Note: anyone feel free to check my math.

 

In any case, there isn't going to be a magic bullet to bring the latency in-line with what you are seeing with HPET disabled. And like I said before, it may not even be accurately reported results in any case. I think this whole thing is largely splitting hairs over something that isn't an issue. If you were seeing large spikes then that would be different story, but you aren't. If you are having performance issues, it has to be elsewhere from DPCs based on your own results.



#102 WonG.

WonG.

    Neowinian

  • Joined: 07-February 14

Posted 11 February 2014 - 01:05

I didn't want to make anyone believe anything. I just asked for some help. If you aren't able to deliver, just don't reply, thanks a lot! That's why a question can't be answered by a countering one. Or do I have to justify before why I'm aiming for a special goal?

 

I really don't care about your maths. You should just be happy for being one of these persons, who aren't sensitive to things like that. I'm not able to see a mistake in your calculation, but maybe you didn't consider every single aspect? Who knows, just an example: Have you ever seen a Film set up for PAL (50Hz) on a 60Hz display? Many people would say, having more Hz/s than needed can't be worse. Of course it can: The asynchronism forces the display to double some frames (nearly every 10th or 11th), this makes the picture unsmooth and micro stuttering appears.

 

I am not the one who sees some numbers and measured "facts", and subsequently my cognition changes... For example, I had to decide bewteen two TVs a few days ago and I took the one representing worse facts than the other one. Believe it or not, it had the better picture. Just for my cognition, it was better than the other one. I don't give a sh... how experts or benchmarks think about this decision, because I have to watch TV on this device, no one else.

 

With my current system settings, I got some issues like micro stuttering and input lag in some cases. I don't care, if i reach 300 or 350fps in a game, but with 300fps or more, it should be displayed smooth, not stuttering... And there are some parallels between these effects and my dpc latency. That's the only reason, why I'm trying to lower it.



#103 Athernar

Athernar

    ?

  • Joined: 15-December 04

Posted 11 February 2014 - 01:22

I didn't want to make anyone believe anything. I just asked for some help. If you aren't able to deliver, just don't reply, thanks a lot! That's why a question can't be answered by a countering one. Or do I have to justify before why I'm aiming for a special goal?

 

I really don't care about your maths. You should just be happy for being one of these persons, who aren't sensitive to things like that. I'm not able to see a mistake in your calculation, but maybe you didn't consider every single aspect? Who knows, just an example: Have you ever seen a Film set up for PAL (50Hz) on a 60Hz display? Many people would say, having more Hz/s than needed can't be worse. Of course it can: The asynchronism forces the display to double some frames (nearly every 10th or 11th), this makes the picture unsmooth and micro stuttering appears.

 

I am not the one who sees some numbers and measured "facts", and subsequently my cognition changes... For example, I had to decide bewteen two TVs a few days ago and I took the one representing worse facts than the other one. Believe it or not, it had the better picture. Just for my cognition, it was better than the other one. I don't give a sh... how experts or benchmarks think about this decision, because I have to watch TV on this device, no one else.

 

With my current system settings, I got some issues like micro stuttering and input lag in some cases. I don't care, if i reach 300 or 350fps in a game, but with 300fps or more, it should be displayed smooth, not stuttering... And there are some parallels between these effects and my dpc latency. That's the only reason, why I'm trying to lower it.

 

You don't care about maths when discussing a piece of hardware that does nothing but maths in the first place? :rolleyes:

 

The placebo effect works both ways, just ask anyone that has had panic/anxiety disorders, if you think hard enough that something is going to happen, your brain will make it happen.

 

This is why we have maths and facts, so we aren't ruled by silly placebo effects and snake oil.



#104 WonG.

WonG.

    Neowinian

  • Joined: 07-February 14

Posted 11 February 2014 - 01:50

Yes, but this maths is way higher than you could ever do this. Not the math is the problem, we are the problem. Often we aren't able to use math so complex, to cover things from every angle we could watch. Maths often only works in models, in a special frame of reference.

 

Or is there any math to display "micro stuttering"? It's like: "You don't need more than 60Hz, your eye isn't able to recognize more than 24 pictures a second..." Unfortunately the math isn't as easy as it seems to be! Anyway, I (and many others) was (were) able to walk through LAN parties and to trace every single display, running in 60Hz mode or less. Not a big deal, but now I accept that you aren't able to recognize micro stuttering and that you aren't able to trust in God. In this case I show a lot more tolerance than you do, not accepting me. And if there is no measurement and no maths for a evidence of micro stuttering, it's just because it hasn't been invented or discovered, yet!

 

Maybe a high speed camera could help, but these are way more efforts, to just get a simple answer, than reasonable. 



#105 +snaphat (Myles Landwehr)

snaphat (Myles Landwehr)

    Electrical & Computer Engineer

  • Tech Issues Solved: 29
  • Joined: 23-August 05
  • OS: Win/Lin/Bsd/Osx
  • Phone: dumb phone

Posted 11 February 2014 - 01:50

I didn't want to make anyone believe anything. I just asked for some help. If you aren't able to deliver, just don't reply, thanks a lot! That's why a question can't be answered by a countering one. Or do I have to justify before why I aim for a special goal?

:rolleyes: This is a forum, it's perfectly valid to ask questions, provide discussion, etc. I've provided help and expert opinion for you in all of my posts up to this point. You don't need to be rude just because I'm disagreeing with you are saying and asked for evidence. 

 

I really don't care about your maths. You should just be happy for being one of these persons, who aren't sensitive to things like that. I'm not able to see a mistake in your calculation, but maybe you didn't consider every single aspect? Who knows, just an example: Have you ever seen a Film set up for PAL (50Hz) on a 60Hz display? Many people would say, having more Hz/s than needed can't be worse. Of course it can: The asynchronism forces the display to double some frames (nearly every 10th or 11th), this makes the picture unsmooth and micro stuttering appears.

This PAL discussion is a digression, but I'm going to assume you are talking about a TV that accepts a PAL signal and then does a poor job converting that to a NTSC signal.

 

In any case, this is neither her nor there w.r.t. to what I said regarding the processing differences between DPC and monitors. Your statement is just: 'you could be wrong because for example some people incorrectly believe that video frame rate conversation can't result in stuttering.' Sure, they are wrong, but people can be wrong about many things.

 

I am not the one who sees some numbers and measured "facts", and subsequently my cognition changes... For example, I had to decide bewteen two TVs a few days ago and I took the one representing worse facts than the other one. Believe it or not, it had the better picture. Just for my cognition, it was better than the other one. I don't give a sh... how experts or benchmarks think about this decision, because I have to watch TV on this device, no one else.

It's fine if you don't care. I'm not here to make you agree with me. I'm just providing comment and help for you because you asked in your posts. Again, these are perfectly normal on a discussion forum.

 

 

With my current system settings, I got some issues like micro stuttering and input lag in some cases. I don't care, if i reach 300 or 350fps in a game, but with 300fps or more, it should be displayed smooth, not stuttering... And there are some parallels between these effects and my dpc latency. That's the only reason, why I'm trying to lower it.

If you are seeing micro-stuttering caused by DPC issues it would be showing up in the graph results for DPC latencies, that's all I'm saying. If the latencies aren't spiking and they are consistent then that can't be the cause of the stuttering.

 

 

Yes, but this maths is way higher than you could ever do this. Not the math is the problem, we are the problem. Often we aren't able to use math so complex, to cover things from every angle we could watch. Maths often only works in models, in a special frame of reference.

 

Or is there any math to display "micro stuttering"? It's like: "You don't need more than 60Hz, your eye isn't able to recognize more than 24 pictures a second..." Unfortunately the math isn't as easy as it seems to be! Anyway, I (and many others) was (were) able to walk through LAN parties and to trace every single display, running in 60Hz mode or less. ... And if there is no measurement and no maths for a evidence of micro stuttering, it's just because it hasn't been invented or discovered, yet!

There is a way to measure micro-stuttering, you graph the frame processing latencies in the same way that we are graphing DPC latencies. If you have consistent latencies then the issues you are having have to be else were, that's just science. :laugh: FYI: Nvidia has a toolkit called FCAT to measure microstuttering, but I've never used it though.

As for eye operating at 24 hertz, that's just a myth, that's not how visual processing in eye/brain works. It's well known that 60 hertz monitors do a poor job of capturing motion blur for example. In any case, this discussion is best done in another thread as Neowin already has had a recent thread regarding this topic. And there are other folk here who are better equipt to comment about it than I am.