Discussion Arrow Lake Builder's thread

Page 7 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

gdansk

Diamond Member
Feb 8, 2011
4,186
7,027
136
And now it is $270 on Amazon for a 265K. Not as good as Newegg's deal but if you don't want ugly gamer memory it may be cheaper. That's a price that I think kills anything but the very high end AMD parts for multi-use DIY PC.

I am worried about Intel. Between this, the B580, and very cheap 226V laptops their consumer margin might be plummeting this year. Or perhaps TSMC is cheaper than claimed.
 
Jul 27, 2020
25,969
17,921
146
Or maybe the refresh CPUs are currently in mass production so they need to clear out inventory. I hope that's the real reason.
 

511

Platinum Member
Jul 12, 2024
2,846
2,841
106
I am worried about Intel. Between this, the B580, and very cheap 226V laptops their consumer margin might be plummeting this year. Or perhaps TSMC is cheaper than claimed.
TSMC is not cheaper than Intel manufacturing same thing in their own fabs also for B580 i worry that Intel will redirect All the dies to AI Cards.
 

511

Platinum Member
Jul 12, 2024
2,846
2,841
106
Yeah this is even more difficult than 18A getting Foundry Customers
 

alcoholbob

Diamond Member
May 24, 2005
6,380
448
126
Is there any plan to push DDR5 speeds over 10K? Like at least 11,000 MT/s? Preferably in 64 or 96GB configurations that don't cost $1000? Only way I can see this platform being slightly interesting.

I assume Nova Lake will support DDR6, which may make it a usable gaming platform if reported speeds of 16,800 MT/s range become commercially available (and assuming AMD choses to gimp itself memory speed-wise again next gen).
 
Jul 27, 2020
25,969
17,921
146
Is there any plan to push DDR5 speeds over 10K? Like at least 11,000 MT/s? Preferably in 64 or 96GB configurations that don't cost $1000? Only way I can see this platform being slightly interesting.
It will be a miracle if the Core 300 series is able to run Gear 2 XMP 128GB DDR5-10000 kit reliably and consistently. No boot failures. Rock stable. If they manage to do that, they deserve an all expenses paid nice vacation to relax.

Current CUDIMM experience is a joke. Blatant lies. People with 9600 kits can barely make it to 9200 Gear 2.
 
  • Like
Reactions: lightmanek

511

Platinum Member
Jul 12, 2024
2,846
2,841
106
Is there any plan to push DDR5 speeds over 10K? Like at least 11,000 MT/s? Preferably in 64 or 96GB configurations that don't cost $1000? Only way I can see this platform being slightly interesting.
DMR Server DDR5 has been planned to be pushed to be upto 12800MT/s using MRDIMMs and Both Zen6 and DMR are supposed to be 16Channel platform.
I assume Nova Lake will support DDR6, which may make it a usable gaming platform if reported speeds of 16,800 MT/s range become commercially available (and assuming AMD choses to gimp itself memory speed-wise again next gen).
NVL is DDR5 8000 JEDEC From what I know with XMP I don't doubt it would be able to do 10K+
It will be a miracle if the Core 300 series is able to run Gear 2 XMP 128GB DDR5-10000 kit reliably and consistently. No boot failures. Rock stable. If they manage to do that, they deserve an all expenses paid nice vacation to relax.

Current CUDIMM experience is a joke. Blatant lies. People with 9600 kits can barely make it to 9200 Gear 2.
It's not that difficult considering the JEDEC Speed is 8000 for NVL if anything the real challenge would be 12K+ Memory Speed.
 
Jul 27, 2020
25,969
17,921
146
NVL is DDR5 8000 JEDEC From what I know with XMP I don't doubt it would be able to do 10K+
That's BORING considering how late NVL is going to be. They should push for 10000 JEDEC, even if it needs to be at CL58 for wider availability and cost reasons.
 
Jul 27, 2020
25,969
17,921
146
I urge you to follow some tuning guides, like the PCGH or Skatterbencher if you haven't already. I got my memory latency down to 77ns, D2D is 3,4, NGU 3,3 and Ring at 4. 4,9E and 5,5P and 7200 UDIMM. It feels like Raptor Lake !
Can you post your Maxxmem2 score, RAM latency score and rudi_float_bench score (all with 7-Max and the latter two benchmarks from my sig) as a comparison point?

What voltages did you use for D2D, NGU and ring?
 

IEC

Elite Member
Super Moderator
Jun 10, 2004
14,595
6,067
136
There have been some good deals recently on Z890/265K/F combos.

I bought a Newegg combo deal for:
Intel 265KF CPU
ASRock Z890 Lightning Mobo
MSI 240mm liquid cooler
and 3 game bundle (SW Outlaws, Civ 7, Dying Light: The Beast)
for $377. Not bad.

Technically -$10 MIR (electronic submit) for the AIO cooler too.

For a modern platform pretty hard to beat.
 
Jul 27, 2020
25,969
17,921
146
I just had a frustrating morning with my 245KF. I have a huge Excel sheet from work that takes about 260 seconds to calculate on a Xeon 6248R (company server) with 48 threads. So I thought it would be a good idea to test how my 245KF does. WELL, initial impressions were great. I got the lowest time of 209 seconds with only 14 available threads for the computation workload. But then in an attempt to better my score with RAM overclocking (it booted up with 5600), I raised the clocks to 8000 which I know it can do easily with my RAM kit. That worked and was stable in y-cruncher so I benchmark the sheet and...the only way to describe what happened next is a whole string of expletives.

In my previous two or three benchmark attempts, all cores were being used and CPU utilization was hitting a full 100%. After the RAM overclock, the system went full wonky and started throttling the CPU utilization in between the computation so that it was fluctuating between 40 to 100% with major dips. I know about clock throttling but this is the first time I've seen a CPU do "utilization throttling" where the whole thread goes to sleep for a second or two, dropping the utilization across the cores. The time to complete came out higher than 300 seconds which was a HUGE regression.

So RAM overclocking is either broken in the BIOS (I'm on stock Gigabyte BIOS) or this CPU cannot work full tilt with RAM overclocking. There's also weird behavior in Win Server 2019 (so Win10 behavior may be similar) where the workload starts on all cores but after a few seconds the Lion Coves are put to sleep. So here's what happened:

Attempt 1: Lion Coves enjoying their nap during the entire run

Attempt 2: Reboot and this time, for about 10 second intervals multiple times during the run, the Lion Coves participated.

Attempt 3: Reboot and my hopes got raised when I saw the Lion Coves holding on for more than 30 seconds. That turned to dismay when they went to sleep after about 60 seconds and the rest of the run continued with the E-cores.

The workload is a simple sumif() of 76000 rows from data of 1 million rows.

My stupid guess is that without the Win2019 scheduler being hybrid aware, the Intel Thread Director is "learning" as it goes so maybe after about a month, it will start trusting the Lion Coves through the entire run. It's still a very weird behavior making me wonder, "What are they so scared of about Lion Coves running on an unsupported Windows Scheduler that they specifically designed logic to forcefully put them out of commission????".

I'm all ears for anyone with similar experiences. I'm in no mood to update the BIOS because I don't want to trade one fix for a myriad of new potential issues.