-
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"
- Thread starter Dave2150
- Start date
Associate
- Joined
- 30 Mar 2017
- Posts
- 865
So weird. I have the Encore and it doesn't do this at stock for me at all.TVB is the same on the Encore, this is completely stock
However, I enabled Overclocking TVB and +2 boost and works a charm, with all the other boxes for Temp and Bin filled out.
x58 all core and x60 two cores setup, giving me a nice little boost and it's fun to see 6.2Ghz at times! Get 6Ghz all core quite a lot in low work loads and easily does 5.9Ghz most of the time all core in many games.
I know there are better results, but these seem to work for me, so pretty happy.
Also, glad I reenabled HT and E cores. I can always use Project Lasso to help with CPU affinity.
Last edited:
These KS's are a bit of a minefield tbh.
They do have 2 issues, one is a VID and the other is a SA. The VID you can sort of work around but not all the time, the SA is more of an issue as it will hard lock on an SA over 1.25.
I've had a few of these now and 2 have had an SA issue, one of them was a very good SP and the other was average but it made using any XMP on higher-end sticks impossible. The VID issue will randomly crash, which can be forced via OCCT or something like that as I have one KS here at the moment and I have just arranged an RMA. As an example of how bad the VID issue can be, on an ASUS MB you can change the SVID to "Intel Fail Safes" which disables MCE and it still crashes or doesn't boot. I also tried that CPU in 3 different boards I have here, 2 x Encores and 1 white Apex same net result, I do have a Dark as well but that thing is a museum piece now
So the 2 KS's which I have here at the moment one is an SP 102 and the other is an SP112 (OCUK) both are from UK stock, and both have very strong IMCs which is good news. That said my 14900K which is from the US is a SP 104 and the other US chip I had would have been very good if it wasn't for the SA bug making it unusable.
These were the CPUs with issues I had:
X346 - SP 103 - SP-P 115 SP-E 81 MC-SP 91 - SA Bug - Hard Lock
X348 - SP 109 - SP-P 123 SP-E 81 MC-SP 88 - SA Bug - Hard Lock
X347 - SP 102 - SP-P 115 SP-E 76 MC-SP 80 - VID Bug - Hard Locks / Intermittent Booting / BSOD - Stress Testing
X346 - SP 103 - SP-P 115 SP-E 81 MC-SP 91 - SA Bug - Hard Lock
X348 - SP 109 - SP-P 123 SP-E 81 MC-SP 88 - SA Bug - Hard Lock
X347 - SP 102 - SP-P 115 SP-E 76 MC-SP 80 - VID Bug - Hard Locks / Intermittent Booting / BSOD - Stress Testing
Last edited:
All the good ones i've seen especially over on overclock.net are from the US. The UK batch seems pants in comparison.
Yep purchased 2 from elsewhere , 1st i put in was SP99 and i felt like crying then the 2nd was SP108 which was much better and kept that one
2 weeks later i decided to delid it and lets just say that didn't go as planned
Quick call to Kris Fix in Germany and after a week was on its way back to me
KS ownership has been a fun experience so far
Last edited:
Associate
Has there been more progress with this whole crashing issue? Im contemplating upgrade and this has thrown me off a fair bit.. Seems only Unreal Engine games though, or is that not correct?
It is only a small number of CPUs affected, largely 13900/14900 though a smaller number of people have reported it on lower models, and in most cases when combined with Asus motherboards, it can usually be fixed by manually setting the boost limits to Intel spec.
Thanks for the detailed explanation. You seem to get lucky with OCUK Cpu. I remember getting a same batch 13900K as you, yours was like SP120 or something and mine came out SP 102, hopefully I'll get a good sample finally. For me though the most important thing is efficiency and strong IMC to run 8400+ which my current 13900KS can do. I intend to run stock under volt at hopefully 8600 38CL ram.
Yes, my 13900K had an SP-P 120 but had fairly weak cores, my 13900KS was a lot better than that one.
Here are my best 14900KS so far which is an SP112 overall, which is an SP-P 123 SP-E 90 MC-SP 85. I am not sure I have seen another one with an E core in the 90's which is interesting but for now the CPU sitting on the side.
What I have noticed with these though, other than the MC-SP being a bit of a meme as it's worked out with SA vs FMAX, MCs around the 84-88 mark seem to be very strong and the ones with 90+ can be interesting as they often have a SA bug and they need a bit more tinkering than the 84-88 mark.
Here is the list of all the 14900KS's I have had to to date:
UK - X401 - SP101 - SP-P 113 SP-E 77 MC-SP 80
UK - X401 - SP102 - SP-P 112 SP-E 82 MC-SP 88
UK - X346 - SP103- SP-P 117 SP-E 77 MC-SP 82
UK - X346 - SP103 - SP-P 115 SP-E 81 MC-SP 91 - SA Bug - Hard Lock - RMA'ed
US - X348 - SP 109 - SP-P 123 SP-E 81 MC-SP 88 - SA Bug - Hard Lock - RMA'ed
UK - X347 - SP 102 - SP-P 115 SP-E 76 MC-SP 80 - VID Bug - Hard Locks / Doesn't Boot / BSOD - - RMA'ed
UK - X343 - SP112 - SP-P 123 SP-E 90 MC-SP 85
I was going to order a couple of others from the US but I am mostly done with them now.
As a general rule, most of my best CPUs have all come from the US.
Last edited:
Yes, my 13900K had an SP-P 120 but had fairly weak cores, my 13900KS was a lot better than that one.
Here are my best 14900KS so far which is an SP112 overall, which is an SP-P 123 SP-E 90 MC-SP 85. I am not sure I have seen another one with an E core in the 90's which is interesting but for now the CPU sitting on the side.
What I have noticed with these though, other than the MC-SP being a bit of a meme as it's worked out with SA vs FMAX, MCs around the 84-88 mark seem to be very strong and the ones with 90+ can be interesting as they often have a SA bug and they need a bit more tinkering than the 84-88 mark.
Here is the list of all the 14900KS's I have had to to date:
UK - X401 - SP101 - SP-P 113 SP-E 77 MC-SP 80
UK - X401 - SP102 - SP-P 112 SP-E 82 MC-SP 88
UK - X346 - SP103- SP-P 117 SP-E 77 MC-SP 82
UK - X346 - SP103 - SP-P 115 SP-E 81 MC-SP 91 - SA Bug - Hard Lock - RMA'ed
US - X348 - SP 109 - SP-P 123 SP-E 81 MC-SP 88 - SA Bug - Hard Lock - RMA'ed
UK - X347 - SP 102 - SP-P 115 SP-E 76 MC-SP 80 - VID Bug - Hard Locks / Doesn't Boot / BSOD - - RMA'ed
UK - X343 - SP112 - SP-P 123 SP-E 90 MC-SP 85
I was going to order a couple of others from the US but I am mostly done with them now.
As a general rule, most of my best CPUs have all come from the US.
How do you test for the SA bug, Vid bug and hard lock? I haven't done much tinkering for awhile. Cheers
Both are fairly easy to test against.
SA Bug - This will normally hard lock/crash on SA voltages over 1.25v
So as an example, using a G.Skill 8200 kit if you enabled XMP on both Asus Z790 Apex MB's will set the SA to around 1.25 and then depending on how bad the bug is on the specific CPU it may not even boot just hard lock on the boot screen. Then you have to start backing down the SA voltage and then it starts to boot and becomes usable. However, then you have the task of balancing the SA vs IMC voltage which then may not be stable at the required frequency.
VID Bug - This will normally crash/lock when pushing the CPU.
So as an example on the CPU which I have just RMA'ed, sometimes it booted and others it didn't. When it booted normally running something like OCCT on a simple testing etc... will crash the PC. In both cases, these are using complete default bios settings. Even changing the SVID option to use the "Intel Fail Safe" (This will also disable MCE) the CPU was crashing on boot. However, you can correct the issues by downclocking the CPU which seemed to work on this CPU and then stress tests ran as normal but also manually setting the voltages in this used case didn't help the issue. Also to add before someone comments, cooling the CPU wasn't an issue and these never really peaked over 80C as I have a fairly beefy cooling system for my bench.
The VID table for those who don't know are the burnt-in values the CPU uses at a given frequency
Both tests are completely repeatable and I have several other CPUs which pass the same tests with zero issues, one is a high bin (SP112) and the other is a lower bin (SP102)
Those seem to be the common things, I have seen with these chips.
Got to be one of the best KS i have seen so far : https://www.overclock.net/threads/o...lts-bins-and-discussion.1807439/post-29320765
Which then ends up the same as mine 
: https://www.overclock.net/threads/o...lts-bins-and-discussion.1807439/post-29320838
Told him to send it to Kris Fix
Which then ends up the same as mine
: https://www.overclock.net/threads/o...lts-bins-and-discussion.1807439/post-29320838Told him to send it to Kris Fix
That's quite a bit more than mine running the same kit, as both of my KS's set G.Skill 8400 kit to around ~1.261
1.35 sounds like an IVR TX voltage for those speeds.
X340 is an early batch, normally the early ones are better than the later ones.
Last edited:
I have tested with a G.Skill 7800 (A-Die) / G.Skill 8200 (M-Die) / G.Skill 8400 (M-Die) kits.
I normally see the 8200/8400 kits seem to crash booting into Windows, the 7800 did boot into Windows but as soon as running a memory test like Karhu / VST etc.... Windows Hard Locks, also note the reset button doesn't work when this happens as well. That behaviour normally goes hand in hand with memory issues. Then if you reduce the SA voltage to 1.24 on 8200/8400 kits it boots into Windows but then hard locks at random or when performing a memory test.
On all 3 RAM kits, one had a working SA voltage of 1.18 and the other had a working SA voltage of 1.1 but then 7800+ wasn't stable with the IMC. It's been a while but I think even DDR5 6000 sets an SA around 1.2 from memory.
For me with my testing setting the SA at 1.25 or more on those two CPU's caused them to hard lock and the reset button not to work.
So to answer your question directly, no I don't believe it's just on high RAM speeds.
Agreed, it only seems to be affecting some X batch CPUs specifically out of Vietnam more so on the highly binned ones.