8 Gb for $7000 also, $200 per extra 8 Gb, and soldered to prevent user upgrades.
I don’t think that Intel or AMD need to worry at all about the new Apple laptops.
8 Gb for $7000 also, $200 per extra 8 Gb, and soldered to prevent user upgrades.
I don’t think that Intel or AMD need to worry at all about the new Apple laptops.
My 10900K could do 4533CL15 / 4600CL16 with 1.28v SA. Completely safe and effortless, 10th gen still remains the strongest Intel IMC to this date for 1:1 memory clocks.
Also 4000 is nothing for 10th gen.
Imagine still being on Sata 1, DDR1, and PCIE 1 or even AGP if this was the case.
Only reason I stuck to Intel is I decided to keep my high end DDR4. If I was building a new system with DDR5 right now, I would be getting the 7800X3D.
Theres some people running Micron B rev configs in either 2x32 or 4x16 which can still do 4300-4400 G1. 4xDR compatibility is also listed at 4400 on current DDR4 mobos.
The problem is that ram went EOL like 3 years ago and you can only get it second hand now. I totally screwed up by not getting 4x16 when it went on EOL sale.
This is actually a lot better for 128 Gb than 4800 DDR5 G2.
Well the fact is that every gen, the 900K launches at the same price as the previous one did.
£580 for a 14900K, £1600 for a 4090.
Same with the rest of the chips as well, you basically get at least a faster chip at the same launch price.
AMD aren’t usually any better for pricing, also nothing to compete in the budget segment against non K I3 and I5 chips.
Y cruncher and Prime 95 stability are not absolute requirements for stability.
I use voltages that bsod in Cinebench for being too low for daily gaming without issue.
A 12400 non K is enough for 4K lol.
Funny fact, higher res = less to no difference from CPU because of GPU bottleneck.
Its likely the gigabyte, the majority of them can’t even run the Intel stock 3200 / 5600 ram speeds without Bsods.
Only reason to upgrade from a 12th gen is if you need a better IMC, and if you did you would do it without needing to ask.
No!
You NEED to touch bios settings on 13th / 14th gen!
This CPU won’t even last a year without degrading if left like that.
The motherboard bioses overvolt these chips for worst case scenarios.
Start off with forcing the stock power limit of the chip, I believe its 180w for the 13600K, actual usage in normal use shouldn’t be much higher than 140w.
Next calibrate the AC_LL, for some boards the value will read like 0.04, on others 40 for the same. The 13600KF I had worked at minimum AC_LL, yours might not.
Start with 0.04 then work down by 0.01 at a time and check it is still boosting in cinebench.
Also get HWinfo64 not HWmonitor.
Actually if its for Gaming Ryzen 7000 3D chips are a lot better now, much more performance at hardly any power usage, check the reviews.
I only stuck with Intel because I wanted to stick to DDR4.
I get 40k at first then it drops to 39k as the throttle kicks in.
I can manage 325w PL on a 420mm AIO, but even with offset it still throttles to 5.6 Ghz.
People getting 40k+ are mostly on water cooling.
My 13600KF did 5.4 all core at just 1.25v when it was new. It degraded though and then needed 1.3v. Rip.
Something not a lot people realize is this depends on your ram.
At all stock bios and DDR4 ram set to 2666, my cinebench temps are low 80s, so gaming would be as you mentioned 65.
With 4300 G1 set, cinebench temps go up to 95c, but gaming is still 72c average, 82 max. the same happens with DDR5 or DDR4 when activating XMP, and likely worse because the auto ram voltages will be set too high on auto.
14700K would also give you Intel’s new APO feature, currently it works in 2 games, but where it does work over 200 FPS uplift has been reported.
A year from now more games will support it, and you won’t have it if you get less than a 14700K.
I found with the latest MSI bios I don’t need to touch the LLC setting, just lite load in advanced CPU configuration.
Change it to manual, set AC loadline to 50 to start with, DC loadline to 110-120, then play around with negative offset voltage.
Then change the AC loadline by -10 / +10 until it is stable. Higher AC loadline = higher offset can be applied without too low vdroop, I ended up needing 60 AC LL on my chip for it to not droop to 1.225v or less, so now the voltage curve is 1.25-1.38 with some negative offset applied, and 325 PL stays under 99c in cinebench.
So the best I’ve managed on my chip now is 325w PL, 97c max in cinebench, 1.25-1.38v with manual offset and LL tweaking, and it throttles to just 5.6.
Thats with a 420mm Arctic AIO, nothing I can do to maintain 5.7 all core at under 99c with my ram OC as well.
Best help would be don’t.
With uncapped PL, the 14900K pulls up to around 350w load at stock in cinebench.
These chips are already pushed to their max, you will more than likely struggle to keep it at even 90c at stock with the ram XMP applied.
Better thing to do is undervolt so that the stock 5.7 all core is maintained more often, and set a power limit to whatever your cooler is capable of in cinebench for sub 95c temps. Find the highest negative offset you can apply and still maintain cinebench stability and leave it on auto volts, for me I can manage -0.06v with everything else at stock.
All I know is speed -
https://i.imgur.com/DnLTTFL.png
Before anyone wants to point out that Optane is better for random read / write, do also show me 8 Tb of Optane storage.