Jump to content



Photo

  • Please log in to reply
155 replies to this topic

#106 BoDEAN

BoDEAN

    REDHead

  • Joined: 04-May 04
  • Location: Chicago, IL
  • OS: Windows 8.1
  • Phone: Iphone

Posted 11 February 2014 - 02:02

My performance is currently fine with my setup. Going to pass on this tweak, but thanks for sharing.




#107 WonG.

WonG.

    Neowinian

  • Joined: 07-February 14

Posted 11 February 2014 - 02:08


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.

 

This micro stuttering doesn't express itself as big spikes. Yes, sometimes there are spikes, and then you really are able to measure them via the DPC latency.

 

But anyway, you aren't able to lower my base latency, so all discussions are unnecessary. Thank you very much for your efforts and sorry for being rude.



#108 Athernar

Athernar

    ?

  • Joined: 15-December 04

Posted 11 February 2014 - 02:09

but now I accept that you aren't able to recognize micro stuttering and that you aren't able to trust in God.

 

Typical placebo cop-out response. Computers are man-made digital devices, they are built upon and built from maths and logic. There is no subjectivity or unknowns at play here, a computer is not like the human body where we've yet to understand it fully.

 

If the math doesn't add up, and you've accounted for all the correct variables, then your "theory" is bunk. Simple as that.



#109 WonG.

WonG.

    Neowinian

  • Joined: 07-February 14

Posted 11 February 2014 - 02:16

@Athernar

 

Y, but math should be used correct. There is no way, a normal fps benchmark is able to help me displaying my issue. If there is 1/8 or 1/16 of a second, where no new pictures are shown, and in the rest of the second, there are enough new pictures shown, you won't be able to see the problem on the basis of a normal benchmark. You can have at least more than 300fps without having a smooth picture.



#110 Athernar

Athernar

    ?

  • Joined: 15-December 04

Posted 11 February 2014 - 02:19

@Athernar

 

Y, but math should be used correct. There is no way, a normal fps benchmark is able to help me displaying my issue. If there is 1/8 or 1/16 of a second, where no new pictures are shown, and in the rest of the second, there are enough new pictures shown, you won't be able to see the problem on the basis of a normal benchmark. You can have at least more than 300fps without having a smooth picture.

 

So benchmark the frame time instead of the FPS aggregate.



#111 NateB1

NateB1

    Neowinian Senior

  • Joined: 09-January 07

Posted 11 February 2014 - 02:22

This is purely anecdotal, but I had a case where I overclocked my CPU by a fairly significant amount several years ago( ~800 Mhz or so on an i7 920), and for a while games would operate in a slowed-down fashion - where the audio would play and the characters in the 3D games would quickly go out of sync with the audio, due to being in slow motion.  It was not a framerate issue - the framerate was fantastic, but anything that required CPU was in slow-motion.   It was very annoying - and what fixed it was enabling HPET in the BIOS and in Windows.  It was a very odd problem.  I've moved to a different CPU/motherboard now, but still overclock by a fairly wide margin, and I've kept HPET enabled to prevent such a scenario from happening again.



#112 +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 - 02:26

 

This micro stuttering doesn't express itself as big spikes. Yes, sometimes there are spikes, and then you really are able to measure them via the DPC latency.

 

But anyway, you aren't able to lower my base latency, so all discussions are unnecessary. Thank you very much for your efforts and sorry for being rude.

 Micro-stuttering by definition is caused by differences in frame processing that show up a frame processing spikes. You aren't looking at that directly though. However, it stands to reason that if DPC latency is consistent and low, the micro-stuttering you are seeing can't be caused by it. If you are seeing stuttering you need to check your frame latencies and see if those show an issue. If neither show an issue, then something else is going on. Discrepancies in frame processing time are going to come out in the wash somewhere.

 

So benchmark the frame time instead of the FPS aggregate.

Yup.

 

@WonG: I modified my previous post slightly. It mentions something called FCAT, try to see if you can use that. You don't need to show us evidence if you don't want to, but it'd be probably be good tool check if you can see issues yourself.

 

 

This is purely anecdotal, but I had a case where I overclocked my CPU by a fairly significant amount several years ago( ~800 Mhz or so on an i7 920), and for a while games would operate in a slowed-down fashion - where the audio would play and the characters in the 3D games would quickly go out of sync with the audio, due to being in slow motion.  It was not a framerate issue - the framerate was fantastic, but anything that required CPU was in slow-motion.   It was very annoying - and what fixed it was enabling HPET in the BIOS and in Windows.  It was a very odd problem.  I've moved to a different CPU/motherboard now, but still overclock by a fairly wide margin, and I've kept HPET enabled to prevent such a scenario from happening again.

It kind of sounds like overheating and the CPU throttling to me. It also kind of sounds like buggy timer code or driver code. You aren't the first person to experience this from googling briefly. I'm chalking it up to bugs somewhere at the moment given that it was fixed via HPET in your case. 



#113 WonG.

WonG.

    Neowinian

  • Joined: 07-February 14

Posted 11 February 2014 - 02:41

@WonG: I modified my previous post slightly. It mentions something called FCAT, try to see if you can use that. You don't need to show us evidence if you don't want to, but it'd be probably be good tool check if you can see issues yourself.

With lower latency I haven't had any stuttering issues... The FCAT seems to need a high speed capture card to work correct. The software based capturing captures before directx and before drivers take action.

 

http://international...wer's_Guide.pdf



#114 +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 - 02:52

With lower latency I haven't had any stuttering issues... The FCAT seems to need a high speed capture card to work correct. The software based capturing captures before directx and before drivers take action.

 

http://international...wer's_Guide.pdf

Try Fraps with frame-times enabled and process it using this app: http://sourceforge.n...enchview/files/ (it explains how). Note: I've never done this so I dunno how well it will work. I wasn't aware fraps was capable of doing frame times (perhaps it is a new?)

 

EDIT: another link for it: https://frapsforum.c...ark-files.2121/



#115 NateB1

NateB1

    Neowinian Senior

  • Joined: 09-January 07

Posted 11 February 2014 - 23:36

 

 

It kind of sounds like overheating and the CPU throttling to me. It also kind of sounds like buggy timer code or driver code. You aren't the first person to experience this from googling briefly. I'm chalking it up to bugs somewhere at the moment given that it was fixed via HPET in your case. 

 

 

Temps were fine - I'm blaming the motherboard.  I experienced random odd occasional stability issues with it I could never quite pin down - memory tests came back fine, it passed CPU/GPU stress tests, etc.



#116 clustaB

clustaB

    Resident One Post Wonder

  • Joined: 13-March 14

Posted 13 March 2014 - 11:17

Hi guys, I could not resist joining this forum to add my two cents. HPET made me go nutz, too.

 

 

HPET on/off making a significant difference is definitely NOT snake oil.

 

 

On my previous Win 7 build (DAW only), disabling HPET in BIOS & OS resulted in a DPC latency of comfortable 2 - 4 micro seconds. Before that, the latency rarely ever dropped below 130 micro seconds, which already has been quite good regarding the 270 which I´ve had before I got rid of my nvidia card and its crappy drivers.

 

Not only there were no more drop outs (at reasonable buffer sizes), the whole system was much more responsive. The negative result was that my ASIO load meter got more "nervous/sensitive". Means that the meter in Cubase jumped up and down more than before when it was slightly higher but also more stable.

 

I´ve dealt quite a while with the HPET thing and have come to the conclusion, that HPET is neither the "best timer" nor the worst. Computers are not interested in having the timer "ring" every 1000 or 976 or whatever when. They have no personal preference and don´t care about the output. Just about the processing itself. So the important thing is to have as many timers as possible as this means more potential moments to generate interrupts.

 

["...interrupts can be missed if the target time has already passed"]  the wiki link

 

Enabling HPET & using "bcdedit..... true" to force Windows to use only one timer seems absurd. More

 

Just think of it as drinking in a bar where two out of three waitresses have been fired and the one who´s left may be extremely punctual at appearing only once per hour. One of the former waitresses (which meanwhile are unemployed and hungry) could have come your way already 45 minutes after your last beer order. That´s when you were supposed to generate that one interrupt to quench your thirst.

 

I am aware that there has to be some kind of down side to it. Otherwise, nVidia would not set their drivers to strictly induce a latency of 255 micro seconds instead of matching it with the PCI Latency Setting found in BIOS. Maybe this allows a more stable framerate (not a faster one!), like slaves on a galley working their paddles to the decent but steady beat of the drummer. Less strokes, but more load per stroke.

 

While its benefits/disadvantages depend on each single person´s needs, its influence on the computer at the bare system level has to be clearly explainable. There is no place for "every setup is different, so this function does whatever its wants to do", computers don´t work this way.

 

For me, disabling it much more serves my needs as constantly keeps my snake oiled.

 

 

Cheers from Germany



#117 DeepSpaceDoge

DeepSpaceDoge

    Neowinian

  • Joined: 20-March 14

Posted 20 March 2014 - 06:55

I have just registerd to say that this did wonders to my old rig

Core 2 Quad Q6600@3Ghz

Asus P5Q-Pro on Intel P45

Win 8.1 64bit

 

I have no option in bios to enable HPET but enabling it in windows with Win 8 specific command "bcdedit /set {current} useplatformclock yes"  increased overall responsivness and disk transfer rates, firefox starts like 2x faster same as many other apps, and what is most imortant to me finally Diablo 3 stopped stuttering like crazy every few seconds without anything special on screen - battle.net forums are flooded with posts about this problem and Blizzard completely ignores them. If I only knew about it since D3 release day!

High precision event timer is in my device manager and WinTimerTester shows 14.31818 MHz

What to say more...Big THANK YOU



#118 PanzerIV

PanzerIV

    Neowinian

  • Joined: 22-March 14

Posted 22 March 2014 - 06:56

Me too I've registered on that board only to comment about my experience. It's a comment from a post on BlurBusters.com that brang me here. I already had "HPET" enabled since day 1 on my new computer and even did since forever on my previous as I always optimize my bios but I never knew that you also had to tweak a setting in Windows for it to really be 100% effective. That's imo super dumb and should be auto-detected and change on it's own without the user's manual intervention. I mostly play first person shooters and I'm a perfectionist with my computer so it needs to always be clean both inside my case and on my drives, the OS and Bios needs to be fully optimized, everything that can be overclocked needs to be to reach the most performance out of what I've paid but yet I would still get some stuttering in Battlefield 4 on Ultra 4xAA at 1080P even though I have ~120fps so I wondered wtf was wrong and out of nowhere it appears that I've found the problem. I'm glad I landed on this forum as otherwise I would have never found it! I ran the WinTimer and it goes right away to 0.99 then after about 180 seconds it stays to 1.0 if anyone wonders and here's a part of my specs below. Oh yea I'm on Windows 8.1 64bit btw.

 

Main specifications of my computer:

Processor: Intel i7 4770K @ 4.20Ghz

Motherboard: MSI Z87-GD65 "LGA1150, Chipset Z87"

Memory: Mushkin Redline 2x4Gb DDR3-2133 "9-11-10-28 1.65v"

Video Cards (SLI): Asus Geforce GTX670 2.0Gb "1188Mhz / 7000Mhz, DirectCU II"
                                  Gigabytes Geforce GTX670 2.0Gb "1202Mhz / 7000Mhz, WindForce"

Sound Card: Creative Soundblaster Zxr

Solid State Drive: Samsung 840 Pro 256Gb "Sata-III"



#119 +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 22 March 2014 - 07:59

What is with all of the overly verbose, yet vague, and generally strange posts on this topic all of a sudden? The first contains false analogies (as well as misunderstands how timers work), the second contains absurd claims of increased I/O performance and ridiculous speedups, the third... I'm not really sure what that one is saying. They are all very strange...



#120 Athernar

Athernar

    ?

  • Joined: 15-December 04

Posted 22 March 2014 - 08:12

What is with all of the overly verbose, yet vague, and generally strange posts on this topic all of a sudden? The first contains false analogies (as well as misunderstands how timers work), the second contains absurd claims of increased I/O performance and ridiculous speedups, the third... I'm not really sure what that one is saying. They are all very strange...

 

Tweakers are like the pseudoscience "community", they drift around the place like tumbleweed claiming to have the quick solution to get free performance or free energy. Snakeoil 2.0, like Web 2.0 but even more useless.

 

You'll also notice similar sorts of behaviour around core parking in Windows, it fixed some performance issues for some people with specific Intel chips & SMT enabled but now people preach it like it's the solution to everything.

 

"When all you have is a hammer, every problem looks like a nail"