Page 5 of 22
Re: Is it true Ryzen has higher input lag than Intel? Is there conclusive data to prove this?
Posted: 13 Jun 2020, 21:51
by deama
BTRY B 529th FA BN wrote: ↑13 Jun 2020, 20:17
For some reason I cannot download the Intel Latency tester.
You have to download the .tar file at the bottom and then unzip it twice.
Re: Is it true Ryzen has higher input lag than Intel? Is there conclusive data to prove this?
Posted: 14 Jun 2020, 06:47
by BTRY B 529th FA BN
.
Re: Is it true Ryzen has higher input lag than Intel? Is there conclusive data to prove this?
Posted: 14 Jun 2020, 10:22
by ptuga
BTRY B 529th FA BN wrote: ↑14 Jun 2020, 06:47
When I click the .tar link my ISP says the page isn't found.
Try another browser. For some reason i can't download it on chrome either.
Also delete mlcdrv.sys, it changes prefetcher settings, that's why my results were different from a similar system.
8700k 5ghz + 16gb 4000mhz 15-16-16
I still don't understand why it closes itself on last test, it's not my bios settings as i've tried with optimized defaults.
Re: Is it true Ryzen has higher input lag than Intel? Is there conclusive data to prove this?
Posted: 14 Jun 2020, 11:16
by RealNC
ptuga wrote: ↑14 Jun 2020, 10:22
I still don't understand why it closes itself on last test, it's not my bios settings as i've tried with optimized defaults.
You need to run these scripts from a command line, not by double-clicking on them.
Re: Is it true Ryzen has higher input lag than Intel? Is there conclusive data to prove this?
Posted: 14 Jun 2020, 11:28
by jorimt
ptuga wrote: ↑14 Jun 2020, 10:22
I still don't understand why it closes itself on last test, it's not my bios settings as i've tried with optimized defaults.
To expound on what RealNC said (assuming you're running Windows 10)...
1. Search for "Command Prompt" in the Windows search bar and select "Run as administrator" under "Open"
2. Open the folder your mlc.exe is in, right click on the address bar in the folder window, and then click "Copy address as text"
3. In your command prompt window type "cd" (without quotes), hit space on your keyboard, and then right click in the prompt window to paste your directory. Finally, hit return/enter on your keyboard.
4. type "mlc" in the prompt window (without quotes), and let it run.
Also, don't disable or delete "mlcdrv.sys," you need it for accurate test numbers:
https://software.intel.com/content/www/ ... ecker.html
HW Prefetcher Control
It is challenging to accurately measure memory latencies on modern Intel processors as they have sophisticated h/w prefetchers. Intel® MLC automatically disables these prefetchers while measuring the latencies and restores them to their previous state on completion. The prefetcher control is exposed through MSR (/content/www/us/en/develop/articles/disclosure-of-hw-prefetcher-control-on-some-intel-processors.html) and MSR access requires root level permission. So, Intel® MLC needs to be run as ‘root’ on Linux. On Windows, we have provided a signed driver that is used for this MSR access. If Intel® MLC can’t be run with root permissions, please consult the readme.pdf that can be found in the download package.
I have really nothing to compare mine against, and am honestly not into the whole benchmarking/overclock scene, but here's mine for reference:
- "Balanced" power mode
- 32GB G.SKILL TridentZ DDR4 @3200MHz (Dual Channel: 14-14-14-34)
- 8700k HT enabled, 4.7GHz auto turbo (4.3GHz sustained on all cores depending on the situation; never bothered to overclock to 5GHz)
Re: Is it true Ryzen has higher input lag than Intel? Is there conclusive data to prove this?
Posted: 14 Jun 2020, 11:54
by BTRY B 529th FA BN
.
Re: Is it true Ryzen has higher input lag than Intel? Is there conclusive data to prove this?
Posted: 14 Jun 2020, 17:39
by jorimt
BTRY B 529th FA BN wrote: ↑14 Jun 2020, 11:54
I've tried Firefox and Microsoft Edge, same thing. My ISP can't find the link after I click Agree from the .tgz file
Try clearing your cache in each browser (or going into private/incognito mode) and attempt again. The Intel website appears to be using javascript to prevent a visitor from downloading the file before they first accept the terms of use, and this caused it to glitch out for me a few times as well before it finally downloaded correctly.
Re: Is it true Ryzen has higher input lag than Intel? Is there conclusive data to prove this?
Posted: 20 Jun 2020, 04:41
by deama
Would reducing RAM timings reduce ryzen's input lag by a reasonable amount, or is it mostly a problem with the CPU itself?
Re: Is it true Ryzen has higher input lag than Intel? Is there conclusive data to prove this?
Posted: 02 Jul 2020, 08:59
by RamenRider
I used to believe Ryzen input lag was caused by poor scheduler optimization by Windows, but actually the Input Lag 100% comes from DWM.
I chased DPC latency for 5 years and it turns out none of that shit mattered; I practically lived on Roach's guide and you can still see me on that thread today.
viewtopic.php?f=10&t=7168
Re: Is it true Ryzen has higher input lag than Intel? Is there conclusive data to prove this?
Posted: 02 Jul 2020, 09:29
by Happyalive
It all matters. From bad windows scheduler to more physical distance between componentes within the cpu to +30ns ram latency (and +200< ns loaded ram latency sometimes (on w10)). I don't understand why you'd say all the input lag comes from dwm when you can literally disable dwm on w7-10, and the difference doesn't suddenly vanish.