Horus-Anhur:

Probably it could reduce latency by a few ns. But not much. Maybe 2-4ns. Just guessing since command rate doesn't have that big impact in memory latency. But there's one thing to consider. Those timings are very lose. Cl40 @ 6400 means an absolute latency of 12.5 ns. For comparison, CL16 @ 3800 gives an absolute latency of 8.4 ns For that kit of 6400 to reach that value, it would need to have a CAS of 27. But this is only one timming. And not even the most important.

user1:

intel added a 1/2 rate mode for the memory controller in rocket-lake, known as gear2(I mistakenly referred to this as geardown mode), so if you are running 2t + gear 2 , you get effective 4t, combined with increased latency from the slower memory controller, while this is mainly used to reduce power consumption on current chips, it can be presumed that it is necessary to run DDR5 at its intended rate on alderlake, barring some-kind of unexpectedly massive improvement to intel's memory controller. here is graph from pcgamer showing the effect of gear 2 on performance in f1 2020 [SPOILER] [/SPOILER] gear2+2t is enough to reduce the performance of a 3200mt/s memory kit, to that of a 2400mt/s kit operating in gear1+t1 its a pretty severe hit to latency.

user1:

https://cdn.mos.cms.futurecdn.net/94DFxAirogXns7msJYSaa3.png

southamptonfc:

The conclusion I draw from that graph is that memory speeds and latencies have zero effect on gaming unless you have a 250hz 1080p monitor.

southamptonfc:

The conclusion I draw from that graph is that memory speeds and latencies have zero effect on gaming unless you have a 250hz 1080p monitor.

vestibule:

I read an article the other day that said: Intel had been stuck on 14nm for the paste 6 years. But new architectures had been released to increase performance and of course more cores for the core wars. But in real terms Intel CPU's over the past 6 years have only increased in performance by 10%. 😱

Horus-Anhur:

We are only looking at average FPS. But a CPU that has a better memory subsystem is able to maintain higher minimum frame rates. And less stuttering. This can result in much smoother gameplay.

southamptonfc:

hmmm, the graph has min. framerates. The slowest RAM still manages min. 198 fps. No stuttering there...

Horus-Anhur:

F1 games are not very hard on the memory subsystem. On other games, memory speed can lower both average and 1% low fps. [youtube=VElMNPXJtuA]

southamptonfc:

That video confirms what I said. "In short, in most applications and games, memory speed has little impact". 1080p then yes, you might see a benefit. But as they say, at "realistic" resoultions, 1440p and above, it makes no difference. DDR3200 is as fast as DDR4000. At 4K DDR2400 is 1FPS slower than DDR4000 😀

Horus-Anhur:

It depends on how much a game and resolution is CPU bound.

southamptonfc:

Do you have any examples where at a realistic resolution, that is the case? Because so far, everything that has been posted here shows there is no benefit to gaming with RAM speeds above ~3200

Horus-Anhur:

But you never argued that you must go above 3200 to get huge frame rates. In fact no one in this thread argued that. You are the one that now, defined an arbitrary speed of 3200, just to defend your position. We were talking about how some settings affect latency, such as Gears, comand rate, etc. And what do you mean about "realistic" resolutions. 1080p is the most used resolution used by PC gamers. More than 68% of Steam users run at 1080p.

Horus-Anhur:

We are only looking at average FPS. But a CPU that has a better memory subsystem is able to maintain higher minimum frame rates. And less stuttering. This can result in much smoother gameplay.

Horus-Anhur:

F1 games are not very hard on the memory subsystem. On other games, memory speed can lower both average and 1% low fps. [youtube=VElMNPXJtuA]

southamptonfc:

OK.... 🙄 Let's recap I pointed out that the graph DOES have min. framerates and contradicts what you said.

southamptonfc:

Then you tried to wriggle out of that by saying: You included a video which directly contradicts what you said.

southamptonfc:

"Realistic" resolutions is quote from the video you posted. Did you actually watch it? 3200 again comes from the video you posted. Now I'm certain you didn't watch it.

Horus-Anhur:

And even that graph shows lower minimum frame rates. You are the one that decided that a difference of 100 fps of average, between the fastest and slowest ram, is nothing. And a 70 fps difference in minimum. That video shows games with performance that scale with memory speed. Both in average and in minimum fps. That video tests with 3 resolutions. 1080p, 1440p and 4k.

southamptonfc:

Do you have any examples where at a realistic resolution, that is the case? Because so far, everything that has been posted here shows there is no benefit to gaming with RAM speeds above ~3200