• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

Intel Core i9-10900 10-core CPU Pictured

Joined
Oct 5, 2017
Messages
595 (0.25/day)
It was done by my friend as a part vendor and they already got hands on LGA1200 motherboards so I trust their numbers.
1582301013966.png
 
Joined
Dec 6, 2016
Messages
113 (0.04/day)
Quick question: what's your ram latency in AIDA64 ram test? (you can do this in trail version)

Do you know that currently most games are RAM latency bounded? Intel's gaming advantage is mostly coming from better IMC that can run 4000+ rams ad have less than 40ns latency.

9900KS is not faster than 8086k in those games as less core = less latency for Intel.

See attached. Yes I've explained that to many AMD fanbois. Games are very latency dependent as well as raw clock speed (and IPC obviously). In my real world experience; I've seen 9900k always beat 8700k/8086k due to same IPC, clocking higher on average, more cores to help balance background programs. The very slight latency advantage 8086k has is real, but small. Ryzen has a much higher latency wall to fix and is a much better argument.
 

Attachments

  • aida64 - bench 3.PNG
    aida64 - bench 3.PNG
    150.6 KB · Views: 241
Joined
Oct 5, 2017
Messages
595 (0.25/day)
See attached. Yes I've explained that to many AMD fanbois. Games are very latency dependent as well as raw clock speed (and IPC obviously). In my real world experience; I've seen 9900k always beat 8700k/8086k due to same IPC, clocking higher on average, more cores to help balance background programs. The very slight latency advantage 8086k has is real, but small. Ryzen has a much higher latency wall to fix and is a much better argument.
AMD's IF latency is much higher yes, but the latency inside a single core and a single CCX is actually slightly lower than Intel's ring bus. This being a 2700X, you can also expect that the 3000 series further improved on this.

1582662479853.png
 
Joined
Dec 6, 2016
Messages
113 (0.04/day)
AMD's IF latency is much higher yes, but the latency inside a single core and a single CCX is actually slightly lower than Intel's ring bus. This being a 2700X, you can also expect that the 3000 series further improved on this.

View attachment 145936

Interesting, thanks for sharing that. Looks like the cross-CCX still needs to be vastly improved though.
 
Joined
Feb 3, 2017
Messages
3,481 (1.32/day)
Processor R5 5600X
Motherboard ASUS ROG STRIX B550-I GAMING
Cooling Alpenföhn Black Ridge
Memory 2*16GB DDR4-2666 VLP @3800
Video Card(s) EVGA Geforce RTX 3080 XC3
Storage 1TB Samsung 970 Pro, 2TB Intel 660p
Display(s) ASUS PG279Q, Eizo EV2736W
Case Dan Cases A4-SFX
Power Supply Corsair SF600
Mouse Corsair Ironclaw Wireless RGB
Keyboard Corsair K60
VR HMD HTC Vive
Cross-CCX latency has a part to play but memory latency is more of a suspect for gaming performance. This is considerably higher for Ryzen 3000 series than Intel's for now and actually a bit higher than Ryzen 2000 series.

Techspot has some directly comparable latency graphs:
 

MikeZTM

New Member
Joined
Oct 25, 2019
Messages
16 (0.01/day)
See attached. Yes I've explained that to many AMD fanbois. Games are very latency dependent as well as raw clock speed (and IPC obviously). In my real world experience; I've seen 9900k always beat 8700k/8086k due to same IPC, clocking higher on average, more cores to help balance background programs. The very slight latency advantage 8086k has is real, but small. Ryzen has a much higher latency wall to fix and is a much better argument.

3600 is no where pushing the IMC of 9900k/8086k. Try 4200 with tight timing and you will see the difference.
Sub 40ns latency is really hard on 9900k but much easier on a 8086k.
 
Top