• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

14th Gen "Raptor Lake Refresh"

Jay-G25 said:
yep afaik you just adjust the offset to your liking but if its getting too hot and starts crashing just back it off a bit , mine hasn't caused me a single issue doing it this way even with the undervolt still in place . once you set the offset you can jump back into bios and the table will be updated with the offset numbers added on :)
Click to expand...
Had a bit of a chance to test some different settings with TVB today. I started by just using +1 boost as it automatically recommended higher temp offsets than +2. I imagine that is due to lower power requirements for not hitting 6.2, and instead targeting 6.1. To try and lower temps I reduced LLC level 6 to level 5, as I needed that for stability previously, but figured it may need less now. I kept working with the temp offsets until I could see 6GHz all core all of the time in the games I play the most and tried reloading all the shaders and running a few benchmarks to retest stability. Seems pretty good so far. Thanks for the suggestions! May not be 100% stable, but will see how I get on. I actually ended up getting a bit crazy on the temp offset and went to +34 up from 44C all core to 78C. Temps have definitely increased but it's not insane! The only thing that seems to bring down the clocks now are the momentary core temp jumps above 78C for a split second. Usually stays in the 50's and 60's.

Might not keep these settings, but please see this mental TVB setting and see what you think:

lrbomxg.jpeg
 
Last edited:
Yeah that offset is on the high side and probably needs a touch more voltage to remain stable as temps increase but good that its sort of working and on the right path :D with mine i can only run the +1 profile as stock / undervolted , if i try the +2 it gets in to windows no problem but wont start any games apart from something like heaven so again will need voltage tweaking to get it stable but im not bothered about that as will just run a manual oc if i want 6.1. Not sure if you have seen it but pretty much followed this guide to the T and worked absolutely perfect on the 13900K and could get the +1 profile running no problem but again +2 needed tweaks which is explained how to do it in this guide : https://www.overclock.net/threads/a...00k-an-overclocking-and-tuning-guide.1801569/
 
Last edited:
Jay-G25 said:
Yeah that offset is on the high side and probably needs a touch more voltage to remain stable as temps increase but good that its sort of working and on the right path :D with mine i can only run the +1 profile as stock / undervolted , if i try the +2 it gets in to windows no problem but wont start any games apart from something like heaven so again will need voltage tweaking to get it stable but im not bothered about that as will just run a manual oc if i want 6.1. Not sure if you have seen it but pretty much followed this guide to the T and worked absolutely perfect on the 13900K and could get the +1 profile running no problem but again +2 needed tweaks which is explained how to do it in this guide : https://www.overclock.net/threads/a...00k-an-overclocking-and-tuning-guide.1801569/
Click to expand...
It's been stable whatever offset I have put on it so far, up to +40. The only problem is the temp increases with the offset and if a temp spike hits anything above the offset temp then one of the cores downclocks. Trying to find a happy middle ground where it hits 6GHz most of the time but doesn't push the temps up too much. I think it shows how it is very different from setting 6Ghz all core as an overclock compared to trying to get TVB to aim for it. Still, further ahead than I was before I think and currently using less LLC and max voltage so that is a positive. This is due to changing from +2 boost to +1 that seemed require higher LLC levels.
 
Last edited:
I haven’t read much into the 14th gen but it looks like my motherboard supports it. I’m currently using 12400F but just thinking ahead of upgrade paths.

Is there much point dropping a 14600K while remaining on DDR4?

I like the idea of just dropping a newer cpu but not sure how much hit it is staying on DDR4
 
Jay-G25 said:
Yeah that offset is on the high side and probably needs a touch more voltage to remain stable as temps increase but good that its sort of working and on the right path :D with mine i can only run the +1 profile as stock / undervolted , if i try the +2 it gets in to windows no problem but wont start any games apart from something like heaven so again will need voltage tweaking to get it stable but im not bothered about that as will just run a manual oc if i want 6.1. Not sure if you have seen it but pretty much followed this guide to the T and worked absolutely perfect on the 13900K and could get the +1 profile running no problem but again +2 needed tweaks which is explained how to do it in this guide : https://www.overclock.net/threads/a...00k-an-overclocking-and-tuning-guide.1801569/
Click to expand...
I had previously read the guide you linked, but I went back and had another look at it. I actually had the first proper hard lock crash I have had since messing with the OCTVB offset. I remember setting a maximum CPU temp of 90C and I actually removed the package temperature threshold of 90C that I previously had, as it didn't seem to act quickly enough. The guide mentioned: When we are on impulse overclock and a single core is suddenly capped, an instability can occur in the entire CPU. Particularly, if it is to do some thermal limitation to the CPU. Using the control below (threshold) we can make the same type of limitation, however when the package temperature reaches this value all cores will be capped. Going to give it one more go today with using offsets (with a less drastic approach.)

Can I ask if have any suggestions on changing any settings such as VF curve at point #10 (still don't completely understand that) or have you tired applying a global adaptive (positive) offset like you may do when undervolting, if you noted any instability?

I think what VF curve at point #10 does is change the voltage supplied to the all core load, if I am understanding that correctly?

Edit: I have been throwing more games and stress tests at it. Passes all the stress tests, but I found a test where it always crashed at a certain loading screen is MSFS 2020. When I backed the offset down to (funnily enough) +20 it worked.

So, as above I feel like it needs a bump in voltage to maintain 6GHz all core above 64C. As I am already around 1.4-1.45v absolute max load at all core in games. I feel as though I may have perhaps hit the limit, although I imagine I could probably hit 6GHz with TVB easier than pushing voltage for standard overclock.
 
Last edited:
Tester buys 100s of 13900k and 14900k to stability test motherboard auto optimized default profiles

They found as little as 20% of the CPUs were truely stable. After loading the new Intel baseline profile, nearly all CPUs became stable, at the loss of performance

wccftech.com

Tester Reveals Only 5 out of 10 Core i9-13900K & 2 out of 10 Core i9-14900K CPUs Stable In Auto Profile, Intel & Board Partners Yet To Determine Cause of Stability Issues

Intel's Core i9-14900K & Core i9-13900K CPUs are largedly affected by instability issues as one tester has revealed by testings 100s of chips.
wccftech.com wccftech.com
 
Last edited:
Grim5 said:
Tester buys 100s of 13900k and 14900k to stability test motherboard auto optimized default profiles

They found as little as 20% of the CPUs were truely stable. After loading the new Intel baseline profile, nearly all CPUs became stable, at the loss of performance

wccftech.com

Tester Reveals Only 5 out of 10 Core i9-13900K & 2 out of 10 Core i9-14900K CPUs Stable In Auto Profile, Intel & Board Partners Yet To Determine Cause of Stability Issues

Intel's Core i9-14900K & Core i9-13900K CPUs are largedly affected by instability issues as one tester has revealed by testings 100s of chips.
wccftech.com wccftech.com
Click to expand...

As things stand I feel increasingly happy with my decision to buy a 14700K heh.
 
Glad I didn't stick with the latest beta BIOS for my Gigabyte Z790 Aorus Master, though I've not had issues myself, but people widely reporting Gigabyte Z790 boards freezing in the BIOS with the beta version.
 
Last edited:
0121danwilliams84 said:
I had previously read the guide you linked, but I went back and had another look at it. I actually had the first proper hard lock crash I have had since messing with the OCTVB offset. I remember setting a maximum CPU temp of 90C and I actually removed the package temperature threshold of 90C that I previously had, as it didn't seem to act quickly enough. The guide mentioned: When we are on impulse overclock and a single core is suddenly capped, an instability can occur in the entire CPU. Particularly, if it is to do some thermal limitation to the CPU. Using the control below (threshold) we can make the same type of limitation, however when the package temperature reaches this value all cores will be capped. Going to give it one more go today with using offsets (with a less drastic approach.)

Can I ask if have any suggestions on changing any settings such as VF curve at point #10 (still don't completely understand that) or have you tired applying a global adaptive (positive) offset like you may do when undervolting, if you noted any instability?

I think what VF curve at point #10 does is change the voltage supplied to the all core load, if I am understanding that correctly?

Edit: I have been throwing more games and stress tests at it. Passes all the stress tests, but I found a test where it always crashed at a certain loading screen is MSFS 2020. When I backed the offset down to (funnily enough) +20 it worked.

So, as above I feel like it needs a bump in voltage to maintain 6GHz all core above 64C. As I am already around 1.4-1.45v absolute max load at all core in games. I feel as though I may have perhaps hit the limit, although I imagine I could probably hit 6GHz with TVB easier than pushing voltage for standard overclock.
Click to expand...
On mine i am just using a negative global offset which for me is working perfectly for +1 boost but if you are crashing at higher frequencies he mentions to boost vfpoints 8/9/10 on the VF tables which he seems to have done 84mv for each which works for him but that is a 13900K albeit a decent one, by the sounds of it you will need a touch more voltage to maintain 6ghz across the board in games and not get it to crash .. try adding a small bit to each vf point mentioned and see if it helps .
 
Jay-G25 said:
On mine i am just using a negative global offset which for me is working perfectly for +1 boost but if you are crashing at higher frequencies he mentions to boost vfpoints 8/9/10 on the VF tables which he seems to have done 84mv for each which works for him but that is a 13900K albeit a decent one, by the sounds of it you will need a touch more voltage to maintain 6ghz across the board in games and not get it to crash .. try adding a small bit to each vf point mentioned and see if it helps .
Click to expand...
Thanks for the response and apologies for all the questions, but it is really helping me to understand in more detail!

I was looking at the P-core overclocking section fairly early in the guide where we uses OCTVB and then seems to just change this:

If you experience any instability with the transients generated by GeekBench and 3DMark, you will need to make a correction to the VF curve at point #10.
Start with a positive displacement of 10mv and increase until the instability disappears.

I see he changes 8/9/10 VF points further in the guide where it seems to get a little more complicated and using voltage groups and so forth.

I am fairly confident with using adaptive voltage that I believe (hopefully) just changes the whole voltage curve rather than the individual point.

I guess I am trying to use the simplest approach and wondered what point 10 actually controlled. I imagine it is the all core voltage with +1 boost applied?

kUVzbUT.jpeg


See above my terrible VID table. His 13900K seems way better than my 14900K. I think that pic was with +2 boost running though and I believe the only difference is point 11 is 6100.
 
Last edited:
Jony27 said:
I haven’t read much into the 14th gen but it looks like my motherboard supports it. I’m currently using 12400F but just thinking ahead of upgrade paths.

Is there much point dropping a 14600K while remaining on DDR4?

I like the idea of just dropping a newer cpu but not sure how much hit it is staying on DDR4
Click to expand...
The CPUs that have the true raptor lake cache config (2MB per P-Core & 4MB per E-Core cluster) are quite a large chunk faster than the CPUs that don't (any 12th gen, 13th gen i5 non-K, 14th gen i5 non-K except 14600), but you'll only notice that difference if you're CPU bound (e.g. playing at a low resolution, or games like esports with a high-end GPU).

Another option is to get the 12600KF, which is around £150, so it might not cost that much if you're offloading the 12400F.

The impact of DDR4 depends on the frequency/latency of your memory, the game and the resolution, but if you believe that you'll benefit from the faster CPU (in other words: you're mainly CPU-bound, or getting choppy gameplay due to 1% lows) then I wouldn't avoid upgrading just because you don't have DDR5.
 
Last edited:
Happily running an overclocked 13600K in my gaming machine. Not felt the need to go to a 'Full Fat' i9 just yet as the machine is perfect for gaming and doesn't suffer from the crashing issues either.
I will get one to play with in due course, but don't feel it's needed just yet surprisingly.
 
Last edited:
Tetras said:
The CPUs that have the true raptor lake cache config (2MB per P-Core & 4MB per E-Core cluster) are quite a large chunk faster than the CPUs that don't (any 12th gen, 13th gen i5 non-K, 14th gen i5 non-K except 14600), but you'll only notice that difference if you're CPU bound (e.g. playing at a low resolution, or games like esports with a high-end GPU).

Another option is to get the 12600KF, which is around £150, so it might not cost that much if you're offloading the 12400F.

The impact of DDR4 depends on the frequency/latency of your memory, the game and the resolution, but if you believe that you'll benefit from the faster CPU (in other words: you're mainly CPU-bound, or getting choppy gameplay due to 1% lows) then I wouldn't avoid upgrading just because you don't have DDR5.
Click to expand...
Thanks for your input.

I’m playing on ultrawide res with 100Hz and haven’t had any issues with being cpu bound as it’s paired with 7800xt. I think I will just wait for now until I do another cpu/mobo/ram upgrade and put the current parts into my old pc.
 
Grim5 said:
14900ks drops as low as 5.1ghz in games on the new Intel profile


Lmao Tim on Twitter on the money

Click to expand...
I would hope that would be so transparent that now credible reviewer would fall for it.

However, much PR money can a $54 billion company splash around?

Currently almost all the media are treating Intel with velvet gloves or keeping stum. E.g. ComputerBase had a news story on 26th of April saying benchmarks to follow... nothing since: total silence despite the news item already saying they had seen some double figure drops in the few benchmarks they had run.

It's almost enough to start believing in conspiracy stories!
 
To be fair been a lot of messing about with the BIOS and new BIOS releases having issues (freezing, instability, etc.), incorrect values and/or misleading naming on profiles, etc. or being pulled immediately after being uploaded for unspecified reasons, etc. etc. with mostly beta releases rather than a final one, so it is kind of hard to get a good point to work from for the reviewers.

I've not paid much attention to the MSI side, but neither Asus or Gigabyte have covered themselves in glory so far with regards to these fixes.
 
Last edited:
pete910 said:
They want them samples from the upcoming releases;)
Click to expand...
Well, that is the traditional means of getting the media to follow the line PR wants.

A Rtings like review site - only ever review retail units bought with their own money - would be nice but with computer hardware having those launch-day reviews simply too important.
 
You must log in or register to reply here.
Back
Top Bottom