Is it true Ryzen has higher input lag than Intel? Is there conclusive data to prove this?

[jorimt]

Are others getting 10ns type latency?

Sorry, "others" as in?

[BTRY B 529th FA BN]

Are others getting 10ns type latency?

Sorry, "others" as in?

From what I heard from recently asking, latency numbers like that (~10ns) are a bug and average latency depending on things ranges from high 40's to 70+ with DDR5.

[jorimt]

From what I heard from recently asking, latency numbers like that (~10ns) are a bug and average latency depending on things ranges from high 40's to 70+ with DDR5.

Source? Context? A lot is measured in ns.

I'm not claiming any numbers, I just tested my Alder Lake build recently using the same version of MLC I tested my Coffee Lake system with years ago, and presented the results anecdotally as part of the broader conversation in this thread.

Whether MLC (be it the version I used or another) is misreading DDR5 is another question. What is undeniable, is that DDR5 latency can't be directly compared to DDR4, the same goes for DDR4 vs DDR3. Some metrics across hardware generations are 1:1 comparable, others are not.

Honestly, I'm not terribly interested myself either way. Again, I've tested average latency on my Coffee Lake and Alder Lake systems in like-for-like games with an LDAT, and they're no higher than a 5 year old Haswell system running on DDR3.

[BTRY B 529th FA BN]

From what I heard from recently asking, latency numbers like that (~10ns) are a bug and average latency depending on things ranges from high 40's to 70+ with DDR5.

Source? Context? A lot is measured in ns.

I'm not claiming any numbers, I just tested my Alder Lake build recently using the same version of MLC I tested my Coffee Lake system with years ago, and presented the results anecdotally as part of the broader conversation in this thread.

Whether MLC (be it the version I used or another) is misreading DDR5 is another question. What is undeniable, is that DDR5 latency can't be directly compared to DDR4, the same goes for DDR4 vs DDR3. Some metrics across hardware generations are 1:1 comparable, others are not.

Honestly, I'm not terribly interested myself either way. Again, I've tested average latency on my Coffee Lake and Alder Lake systems in like-for-like games with an LDAT, and they're no higher than a 5 year old Haswell system running on DDR3.

Just trying to find the truth, that's why I asked if others are also getting ~10ns latency. I almost jumped ship to a 12900K from your post with those scores; so I'm gonna ask.

[jorimt]

Just trying to find the truth, that's why I asked if others are also getting ~10ns latency. I almost jumped ship to a 12900K from your post with those scores; so I'm gonna ask.

Gotcha, but lower memory latency is mainly only going to increase your minimum framerates where gaming is directly concerned, which, if paired with a capable CPU, will decrease the amount and duration of frametime spikes, and raise your overall average framerate, but we're usually only talking 3-5% improvement in the absolute best of cases.

I.E. there's no miracle cure. It's a mix of all your components, and if we're talking online games, your ISP and the game's netcode (including match-to-match servers, even) will primarily determine the quality and consistency of your experience, sometimes regardless of how OP your hardware is.

Software bottlenecks are heavily underestimated, especially on these forums as of late.

[BTRY B 529th FA BN]

@jorimt Thank you for your time,

Cheers

[jorimt]

@jorimt Thank you for your time,

Cheers

No worries.

[gameinn]

Wow. What a thread.

People basically claiming the reason they aren't a pro gamer is because they have an AMD CPU and not an Intel CPU.

Even though you know Gamers Nexus literally did a whole video completely dispelling this at every level possible that it was run to run variance. Not to mention that testing was done in mid 2020 when 5000 wasn't out which ended up taking like an extra ~8 ns off latency in aida64 by the improved design. Add in improvements to AGESA and Windows scheduler/power plan for Ryzen since 2020 I wouldn't be surprised if AMD actually had less input lag but again this literally doesn't matter.

I've heard some moronic things in my time. One being "What was invented first? Water or Dr. Pepper?" but claiming you aren't at the skill level you should be at a video game because of a CPU is on another level.



This is the temporary desk setup of an Overwatch league player during quarantine. Playing like this didn't hinder his ability at all. He would still win games. Hopefully this ends whatever debate there is.

You're bad because you're bad. What CPU you have won't change that.

[A Solid lad]

Wow. What a thread.

People basically claiming the reason they aren't a pro gamer is because they have an AMD CPU and not an Intel CPU.

..............

No one claimed any of that in this thread, at least not to my knowledge.
We're simply discussing an input lag related topic that is interesting for us. (As you're supposed to on this forum.)

On any other platform, this discussion couldn't have taken place, and people wouldn't have became smarter by participating,
exactly because of people with attitudes like yours, who instantly make the askee feel bad for their question.
Simultaneously deterring anyone from trying to answer/add to the conversation, because no one likes to be made made fun of/taken for a fool.
I've seen this happen countless times on other forums/subreddits/discords/etc.

We don't care if you think this discussion is pointless.
You're free to close this thread and read another one which interests you... but don't ruin this place for everyone else.

The most infuriating thing on any forum I've ever came across is when one asks a question, and other people
(who most of the time know very little about the given topic and as such, couldn't give a satisfactory answer, even if they wanted)
just reply with:

A, "It's a stupid question."
B, "Why are you asking? Your problem isn't this, but rather that."
C, "It doesn't matter, go think about something else."
...or the like.

Part of why you're doing it could be that you're afraid of admitting your ignorance to yourself?
So you try to make it seem like it's information not even worth knowing?
Where would the world be if everyone thought like this...
Whatever the case may be:

Answers/attitude like this, goes completely against what a forum is supposed to be there for.
Discussion with a capital D.

BlurBusters is the exception.
Over here, questions get actual answers.
...or at least they don't get shut down.
That's one of the main reasons why I value this site, and I'd like it to remain this way.

[Chief Blur Buster]

Answers/attitude like this, goes completely against what a forum is supposed to be there for.

Although I appreciate your intent, if possible, please leave the moderating to us. Long forum-warning posts by non-moderators like yours is like people complaining about ReplyAll using ReplyAll in a ReplyAll accident (chain reaction) -- piling noise on top of a mountain of topic distraction.

For the future, please silently click the exclamation mark button at the top of the post:


The above button sends a broadcast notification automatically to the entire moderator team.

...

Wow. What a thread.

People basically claiming the reason they aren't a pro gamer is because they have an AMD CPU and not an Intel CPU.

Even though you know Gamers Nexus literally did a whole video completely dispelling this at every level possible that it was run to run variance. Not to mention that testing was done in mid 2020 when 5000 wasn't out which ended up taking like an extra ~8 ns off latency in aida64 by the improved design. Add in improvements to AGESA and Windows scheduler/power plan for Ryzen since 2020 I wouldn't be surprised if AMD actually had less input lag but again this literally doesn't matter.

I've heard some moronic things in my time. One being "What was invented first? Water or Dr. Pepper?" but claiming you aren't at the skill level you should be at a video game because of a CPU is on another level.

[image]

This is the temporary desk setup of an Overwatch league player during quarantine. Playing like this didn't hinder his ability at all. He would still win games. Hopefully this ends whatever debate there is.

You're bad because you're bad. What CPU you have won't change that.

This was reported by a few users (exclamation mark button) as a forum rule violation. So I am compelled to post a reminder to everyone:



Whatever useful advice everyone here gives, please make sure tone / politeness / grammar / etc is re-worded to meet forum rules. We are not Twitch Chat, Discord Chat, Twitter or Facebook here -- moderation around here is stricter.

Wrapping useful information (e.g. pro players don't need complex rigs, skill can compensate for AMD-vs-Intel deficiencies) with rudeness isn't the way to approach relaying information that this thread may be genuinely out of date. We aren't aiming to be a social media sewer copycat here.

Even if threads are out of date, any continuance of outdated threads needs to be done in accordance to forum rules, like an informative post (ignoring previous posts) with no negative words. Your post has useful content that is contaminated by a tone that we don't welcome around here in these Moderated forums.

Appreciated everyone, and thank you.