Experimental low latency gaming build

Everything about latency. This section is mainly user/consumer discussion. (Peer-reviewed scientific discussion should go in Laboratory section). Tips, 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.
PunksOnHope
Posts: 36
Joined: 18 Jan 2023, 07:47

Experimental low latency gaming build

Post by PunksOnHope » 21 Jul 2023, 17:47

So I got burned with the 5800x3D and 6700xt.
I gamed better with a 2200g apu at 100fps.

I am contemplating building a new pc with the 5700g and overclocking the crap out of it since it's only 170$ now and trying to get 4800mhz mem clk and 2400 fclk. I understand the best I can hope for is 200fps realistically in competitive shooters at 1080p with fsr upscaling but I I honestly think the 5800x3d is the thing that is frustrating me the most playing shooters.

any thoughts on this? I don't have the money for an Intel build and I want to experiment seeing if having the gpu on cpu die results in less latency because im pretty sure apus have hardware accelerated gpu scheduling built in.

I might take out my 2200g pc and test it out again.

babooz99
Posts: 62
Joined: 14 Mar 2023, 11:12

Re: Experimental low latency gaming build

Post by babooz99 » 21 Jul 2023, 18:46

whats wrong with the 5800x3d

PunksOnHope
Posts: 36
Joined: 18 Jan 2023, 07:47

Re: Experimental low latency gaming build

Post by PunksOnHope » 21 Jul 2023, 21:46

babooz99 wrote:
21 Jul 2023, 18:46
whats wrong with the 5800x3d
Latency and mouse feel. It's a chiplet design with 3d cache and the l3 penalty is massive. I may just wait for strix point. 5700g realistically seems like the performance would still be underwhelming.

TheKelz
Posts: 140
Joined: 15 Aug 2022, 17:15

Re: Experimental low latency gaming build

Post by TheKelz » 22 Jul 2023, 05:03

I have i9 10850k, previously 4790k and I also have (and had) this heavy mouse/like on ice issue. Don't think 5800x3d is to blame here, but who knows, maybe it has even worse latency than these.

User avatar
axaro1
Posts: 628
Joined: 23 Apr 2020, 12:00
Location: Milan, Italy

Re: Experimental low latency gaming build

Post by axaro1 » 22 Jul 2023, 05:09

I have a 5800x3d and my system feels highly responsive, I don't know if it can help you but this is what I'm running:
  • C-States DISABLED (c-states can affect USB behaviour on AM4 motherboards, it's honestly better to disable it most of the times)
  • Cool&Quiet DISABLED
  • CPPC ENABLED
  • CPPC Preferred Core DISABLED (this is a must with x3d chips for better workload allocations)
  • PCIe ASPM mode DISABLED (remove PCIE power saving)
  • L1/L2 Stream HW Prefetcher ENABLED
SMU Common Options:
  • APBDIS 1 with Fixed SOC Pstate 0 (forces infinity fabric and memory controller to run at full power)
  • DF Cstates DISABLED (this way you your infinity fabric never run into idle lower power states)
  • NBIO LCLK DPM -> NBIO DPM Control set it Manual to 2 (forces your socket to always run at the highest power states, it doesn't only affect PCIe but it affects the entire socket)
I assume that you are already using CO to make your CPU boost more consistently without thermal throttling.
Also, don't forget that AM4 motherboards have USB ports handled by the chipset and USB ports handled by the CPU, use CPU USB ports (your motherboard specification should look like this: https://i.imgur.com/bfz4yOn.png).

Last but not least, my 5800x3d fails to boost 1-2 cores to 4550mhz, which should happen during ST workloads. For this reason I'm running 102mhz BCLK so every single core boost to 4539mhz, if your mobo has an external clock generator for the CPU you can even reach up top 4700mhz. With 102mhz blck I also get slightly better M.2/SATA/PCIE performance, there's a lot of people arguing that you should only use 100.00mhz but personally I never encountered performance degradation. I've also seen dumb statements about BCLK affecting latency and jitters which is a complete misunderstanding of what BCLK actually does. At most it will affect a bunch of prescalers in a negative way, like in the case of SATA drives not being detected if BCLK is too high, but it will not cause asynchronicity in timed operations because hardware interrupts don't operate based on system clock.
XL2566K* | XV252QF* | LG C1* | HP OMEN X 25 | XL2546K | VG259QM | XG2402 | LS24F350[RIP]
*= currently owned



MONITOR: XL2566K custom VT: https://i.imgur.com/ylYkuLf.png
CPU: 5800x3d 102mhz BCLK
GPU: 3080FE undervolted
RAM: https://i.imgur.com/iwmraZB.png
MOUSE: Endgame Gear OP1 8k
KEYBOARD: Wooting 60he

LAGenjoyer666
Posts: 10
Joined: 20 Jun 2023, 14:52

Re: Experimental low latency gaming build

Post by LAGenjoyer666 » 22 Jul 2023, 09:01

axaro1 wrote:
22 Jul 2023, 05:09
I have a 5800x3d and my system feels highly responsive, I don't know if it can help you but this is what I'm running:
  • C-States DISABLED (c-states can affect USB behaviour on AM4 motherboards, it's honestly better to disable it most of the times)
  • Cool&Quiet DISABLED
  • CPPC ENABLED
  • CPPC Preferred Core DISABLED (this is a must with x3d chips for better workload allocations)
  • PCIe ASPM mode DISABLED (remove PCIE power saving)
  • L1/L2 Stream HW Prefetcher ENABLED
SMU Common Options:
  • APBDIS 1 with Fixed SOC Pstate 0 (forces infinity fabric and memory controller to run at full power)
  • DF Cstates DISABLED (this way you your infinity fabric never run into idle lower power states)
  • NBIO LCLK DPM -> NBIO DPM Control set it Manual to 2 (forces your socket to always run at the highest power states, it doesn't only affect PCIe but it affects the entire socket)
I assume that you are already using CO to make your CPU boost more consistently without thermal throttling.
Also, don't forget that AM4 motherboards have USB ports handled by the chipset and USB ports handled by the CPU, use CPU USB ports (your motherboard specification should look like this: https://i.imgur.com/bfz4yOn.png).

Last but not least, my 5800x3d fails to boost 1-2 cores to 4550mhz, which should happen during ST workloads. For this reason I'm running 102mhz BCLK so every single core boost to 4539mhz, if your mobo has an external clock generator for the CPU you can even reach up top 4700mhz. With 102mhz blck I also get slightly better M.2/SATA/PCIE performance, there's a lot of people arguing that you should only use 100.00mhz but personally I never encountered performance degradation. I've also seen dumb statements about BCLK affecting latency and jitters which is a complete misunderstanding of what BCLK actually does. At most it will affect a bunch of prescalers in a negative way, like in the case of SATA drives not being detected if BCLK is too high, but it will not cause asynchronicity in timed operations because hardware interrupts don't operate based on system clock.

Do you think these are good settings for 5900x too?

PunksOnHope
Posts: 36
Joined: 18 Jan 2023, 07:47

Re: Experimental low latency gaming build

Post by PunksOnHope » 22 Jul 2023, 10:58

axaro1 wrote:
22 Jul 2023, 05:09
I have a 5800x3d and my system feels highly responsive, I don't know if it can help you but this is what I'm running:
  • C-States DISABLED (c-states can affect USB behaviour on AM4 motherboards, it's honestly better to disable it most of the times)
  • Cool&Quiet DISABLED
  • CPPC ENABLED
  • CPPC Preferred Core DISABLED (this is a must with x3d chips for better workload allocations)
  • PCIe ASPM mode DISABLED (remove PCIE power saving)
  • L1/L2 Stream HW Prefetcher ENABLED
SMU Common Options:
  • APBDIS 1 with Fixed SOC Pstate 0 (forces infinity fabric and memory controller to run at full power)
  • DF Cstates DISABLED (this way you your infinity fabric never run into idle lower power states)
  • NBIO LCLK DPM -> NBIO DPM Control set it Manual to 2 (forces your socket to always run at the highest power states, it doesn't only affect PCIe but it affects the entire socket)
I assume that you are already using CO to make your CPU boost more consistently without thermal throttling.
Also, don't forget that AM4 motherboards have USB ports handled by the chipset and USB ports handled by the CPU, use CPU USB ports (your motherboard specification should look like this: https://i.imgur.com/bfz4yOn.png).

Last but not least, my 5800x3d fails to boost 1-2 cores to 4550mhz, which should happen during ST workloads. For this reason I'm running 102mhz BCLK so every single core boost to 4539mhz, if your mobo has an external clock generator for the CPU you can even reach up top 4700mhz. With 102mhz blck I also get slightly better M.2/SATA/PCIE performance, there's a lot of people arguing that you should only use 100.00mhz but personally I never encountered performance degradation. I've also seen dumb statements about BCLK affecting latency and jitters which is a complete misunderstanding of what BCLK actually does. At most it will affect a bunch of prescalers in a negative way, like in the case of SATA drives not being detected if BCLK is too high, but it will not cause asynchronicity in timed operations because hardware interrupts don't operate based on system clock.

I'm currently swapping out memory sticks because I feel it's unstable with 4 dimms. Do u think above 3200mhz matters for vcache?

PunksOnHope
Posts: 36
Joined: 18 Jan 2023, 07:47

Re: Experimental low latency gaming build

Post by PunksOnHope » 22 Jul 2023, 11:13

I'm trying these bios tweaks now and testing

PunksOnHope
Posts: 36
Joined: 18 Jan 2023, 07:47

Re: Experimental low latency gaming build

Post by PunksOnHope » 22 Jul 2023, 11:39

axaro1 wrote:
22 Jul 2023, 05:09
I have a 5800x3d and my system feels highly responsive, I don't know if it can help you but this is what I'm running:
  • C-States DISABLED (c-states can affect USB behaviour on AM4 motherboards, it's honestly better to disable it most of the times)
  • Cool&Quiet DISABLED
  • CPPC ENABLED
  • CPPC Preferred Core DISABLED (this is a must with x3d chips for better workload allocations)
  • PCIe ASPM mode DISABLED (remove PCIE power saving)
  • L1/L2 Stream HW Prefetcher ENABLED
SMU Common Options:
  • APBDIS 1 with Fixed SOC Pstate 0 (forces infinity fabric and memory controller to run at full power)
  • DF Cstates DISABLED (this way you your infinity fabric never run into idle lower power states)
  • NBIO LCLK DPM -> NBIO DPM Control set it Manual to 2 (forces your socket to always run at the highest power states, it doesn't only affect PCIe but it affects the entire socket)
I assume that you are already using CO to make your CPU boost more consistently without thermal throttling.
Also, don't forget that AM4 motherboards have USB ports handled by the chipset and USB ports handled by the CPU, use CPU USB ports (your motherboard specification should look like this: https://i.imgur.com/bfz4yOn.png).

Last but not least, my 5800x3d fails to boost 1-2 cores to 4550mhz, which should happen during ST workloads. For this reason I'm running 102mhz BCLK so every single core boost to 4539mhz, if your mobo has an external clock generator for the CPU you can even reach up top 4700mhz. With 102mhz blck I also get slightly better M.2/SATA/PCIE performance, there's a lot of people arguing that you should only use 100.00mhz but personally I never encountered performance degradation. I've also seen dumb statements about BCLK affecting latency and jitters which is a complete misunderstanding of what BCLK actually does. At most it will affect a bunch of prescalers in a negative way, like in the case of SATA drives not being detected if BCLK is too high, but it will not cause asynchronicity in timed operations because hardware interrupts don't operate based on system clock.

I tried all of these tweaks and I have a 60 mhz drop in prime95 small ffts along with a 4 c heat increase. This then stabilizes to 100mhz below my bios tweaks without them which is just CO -30

PunksOnHope
Posts: 36
Joined: 18 Jan 2023, 07:47

Re: Experimental low latency gaming build

Post by PunksOnHope » 22 Jul 2023, 11:54

101 base clock causes cpu to kill turbo functionality and locks it to 3.43 mhz. I guess if u do base clock u have to adjust multipliers? I thought they were locked for x3d

Post Reply