enhanced turbo, speed shift, C-state ....... on or off? 14900KS

Scuzzmuffin

New member
Joined
Oct 24, 2024
Messages
15
Ive read a bunch of guides around overlocking/undervolting the 14900ks, but there are 3 settings I'm still confused if they should be on or off for best performance, and how they effect overclocking.

1) Enhanced Turbo - should this be on or off?
- Default appears to be on. Ive seen several people leave it alone and a few turn it off,
2) Intel Speed Shift - should this be on or off
- Default appears to be on. Ive only seen it discussed like once, but didnt say whether to keep it on or off
3) C-State - should this be on or off
- Default appears to be on. Ive seen about half say turn off, and half say leave it alone.
4) CPU Switching Freq - Keep on auto or set value, 500, 600, 700, 800?
- Ive seen a few comment places that a value between 600-800 can improve performance, but with no real support on what it does.
5) Turbo Boost Max Tech 3.0 - does disabling this improve performance with overclocking?
- i understand it try to push certain workloads to specific cores, which is suppose to improve performance, but if we are running all core at 57 or 58x should this be turned off?

My goal is to get max performance out of the 14900ks at 58x.
* im currently at 57x with the following results
-----Intel XTU benchmark : 14200
-----Cinebench R23 : 41200
-----Time Spy CPU Test : 23500
***100% load temp of 89c***

Thanks in advance
 
Last edited:
You're right about CB23 needing voltage, I lower it to the lowest stable voltage, if I raise it a bit, my frequency goes higher. Maybe thats why I end up using 5,3ghz, because the voltage is "too low".
BuildZoid thinks that Raptor Lake has some kind of built in CEP-lite that might explain why frequency drops rather that the CPU crashing when voltage is too low - to a point. What you are observing might be evidence of what BuildZoid theorized. I don’t think you have CEP enabled, but what about the CPU Under Voltage Protection? I wonder if that setting does anything post POST?
One thing about CB R23, it needs voltage for high scores.
You’re not the first person to say that. I guess I need to pay attention to that observation. We already know from past observations that Cinebench R23 is a very different workload compared to Prime95 Small FFTs, i.e. Cinebench likes to flirt with high frequencies. And we have already surmised that we need adequate voltage to get those high frequencies. No doubt the flatter loadline (e.g. LLC 3 - 5) helps keep the voltage up.

So let me think this through for a second. There really isn’t any magic to getting high R23 scores. What you need is adequate voltage combined with sky-high PL1/PL2 and potentially IccMax limits. [Scuzzmuffin’s PL1/PL2 are set to unlimited. No doubt his IccMax is also high.] This would lead me to conclude that the real issue people face when shooting for the high R23 scores (being that the aforementioned parameters are adjustable) is actually their cooling solution, which is not easily adjustable. [Scuzzmuffin is only hitting 93C at 404Watts for some reason.] So ultimately our formula = Silicon Lottery + Cooling Solution $$$ = Your Peak Score Limit. [Just like it’s always been.]

What might be adding to the confusion is that CEP-lite thing…because in the old days R23 would either give you a high score or it would crash. There was no in between.
 
Last edited:
I don’t think you have CEP enabled, but what about the CPU Under Voltage Protection? I wonder if that setting does anything post POST?
I have CEP enabled and undervolt protection disabled. Then again, I don't mind at all. It's a speedy chip with too much headache, as long as it hits the 5,7ghz at med load (1,34 volt-ish), I won't mind at all. During heavy load, it runs at 1,17 volt, but 5,3ghz. This only happens in benchmarks and OCCT/Prime95, but not in renders and games, so I'ts a moot point for me
 
I have CEP enabled and undervolt protection disabled. Then again, I don't mind at all. It's a speedy chip with too much headache, as long as it hits the 5,7ghz at med load (1,34 volt-ish), I won't mind at all. During heavy load, it runs at 1,17 volt, but 5,3ghz. This only happens in benchmarks and OCCT/Prime95, but not in renders and games, so I'ts a moot point for me
Okay. CEP=Enabled, CUVP=disabled. Got it!

Yeah, I agree with you that tuning for the highest R23 scores will ultimately lead you down a potential dangerous path, i.e. ignoring Intel’s power limits advice. I always advocate for having a saved OC Profile for when bragging rights are important, but having a game-tuned profile for daily use.
 
Last edited:
Okay. CEP=Enabled, CUVP=disabled. Got it!

Yeah, I agree with you that tuning for the highest R23 scores will ultimately lead you down a potential dangerous path, i.e. ignoring Intel’s power limits advice. I always advocate for having a saved OC Profile for when bragging rights are important, but having a game-tuned profile for daily use.

Definitely, I have my 24/7 profile with HT off, CEP on, undervolted, set multipliers, etc., and a couple 'benching' profiles that I using only sparingly.
 
i will say the liquid metal lowered the temps about 3c across the board. i also use a contact plate and the "ryujin 3 360" aio cooler.
i run the MSI unlimited power setting. how i have my cooling now, seems solid.
my house is at 70f
my cpu idles at 30c on the dot

yeah i think my time spy is being limited by something, but not sure what.
 

Attachments

  • IMG_8236.jpeg
    IMG_8236.jpeg
    1.4 MB · Views: 31
Last edited:
Hi, I apologize for my language, I use a translator.
I have an MSI Z790 Carbon and 14900KS, an Arctik Freezer 3 420.
It was possible to gain a little more than 40,000 in R23. Could you share the BIOS settings?
  • LLC
  • LiteLoad
  • Offset
  • P-E core frequency
  • Ring frequency
  • TVB, CEP
llc - 5
liteload - advance 15/15
offset - -0.085
freq - 58x 46x
ring - 50x
tvb,cep - off
 
i will say the liquid metal lowered the temps about 3c across the board. i also use a contact plate and the "ryujin 3 360" aio cooler.
Good info. Thanks!

yeah i think my time spy is being limited by something, but not sure what.
It’s quite unlikely that this would account for the big difference between your two scores but we do know that TS is sensitive to memory bandwidth and latency. Just an intial thought. Temps could also be a factor once the GPU heat gets dumped into the case and starts to rise. Do you guys have the AIO in the same configuration, i.e. top vs front mounted?
 
Мой 14900KS - Информация об оборудовании
llc - 5
liteload - advance 15/15
offset - -0.065 (0.085 unstable, crashes in games)
freq - 57x 45x Без имени.jpg
tvb,cep - off
 
Last edited:
Back
Top