Pcb code can guarantee nothing. I doubt you can find your golden sample from 100 cpus. what is you want to achieve? 1.2V 4.5G?
ASUS Rampage IV Black Edition: The Ultimate LGA2011 Motherboard
- Thread starter Kaapstad
- Start date
Soldato
If the voltages are similar to sandy-e i was fine running that at 1.4v with the cooling i had and i'll have about the same on tap this time.
Tbh 4.4/4.5 will be plenty for 24/7
Tbh 4.4/4.5 will be plenty for 24/7
Do you have a source on that? Cause ive seen it for my own eyes.
Actually my brother has a 4930k that does 5ghz on water Ht-on at 1.35v. Its stable in Prime 95,occt and linx.
He also bought that cpu from ocuk, it was his 3rd 4930k. Go ask 8pack and see what he says....he probably wont tell because its supposed to be a secret!
Here is his table of results from a previous cpu.
PCB code does mean something, you seen it here first!
Last edited:
Soldato
Presuming the pcb code is like a grid reference on the wafer (batch)??
Associate
Nick dont go stealing my good chip! ;p lol
I dont know exactly what it is, ill ask my bro, but he swears by it and hes been buying cpus and computers since the early 1980's to the present day. Since the beginning literally
He knows his stuff.
Do you have a source on that? Cause ive seen it for my own eyes.
Actually my brother has a 4930k that does 5ghz on water Ht-on at 1.35v. Its stable in Prime 95,occt and linx.
He also bought that cpu from ocuk, it was his 3rd 4930k. Go ask 8pack and see what he says....he probably wont tell because its supposed to be a secret!
Here is his table of results from a previous cpu.
PCB code does mean something, you seen it here first!
Discussing this is pointless,several questions:
1) how to define "stable'? In my opinion, the definition of stability is rather subjective. Did your brother run all three Prime 95 tests for 12+ hrs? Did he use the latest version of P95 with AVX?
2) I don't any source for that, it's just my own theory based on the statistics from Raja. He is ASUS engineer and did some binning work of 4960x personally.
Also based on some info. I gathered from overclock.net. I don't deny your brother is lucky enough to grab a golden chip. I just thought statistics is more convincing because these statistics are based on tremendous number of samples.
you need to be aware that CPU OC only, RAM OC only, and OC CPU and RAM at the same are 3 different scenarios. Required Vcore is therefore quite different. My point is it's unfair and pointless to just simply compare the volts with others posted online. Maybe your brother is right, but his theory is only based on very limited number of samples. It's all about probability.
Just curious, how did your brother deal with these 4930k that not clock very well? return them to OCUK?
Last edited:
He sold them on, ill be getting his Golden one when he upgrades. One of the reasons why i want the r4eb so late in to the 2011 socket cause ill get his golden cpu.
Im just waiting for the damn mobo to arrive, for the time being ill put in a 4820k and oc it as much as i can.
I already have a 900d, 3xRx360 rads, 2xD5 Vario pumps and the best Koolance reservoir money can buy.
Just need the damn board!!
Last edited:
Man of Honour
Getting tired of seeing pictures of delivered boxes. 
OcUK Staff
Discussing this is pointless,several questions:
1) how to define "stable'? In my opinion, the definition of stability is rather subjective. Did your brother run all three Prime 95 tests for 12+ hrs? Did he use the latest version of P95 with AVX?
2) I don't any source for that, it's just my own theory based on the statistics from Raja. He is ASUS engineer and did some binning work of 4960x personally.
Also based on some info. I gathered from overclock.net. I don't deny your brother is lucky enough to grab a golden chip. I just thought statistics is more convincing because these statistics are based on tremendous number of samples.
you need to be aware that CPU OC only, RAM OC only, and OC CPU and RAM at the same are 3 different scenarios. Required Vcore is therefore quite different. My point is it's unfair and pointless to just simply compare the volts with others posted online. Maybe your brother is right, but his theory is only based on very limited number of samples. It's all about probability.
Just curious, how did your brother deal with these 4930k that not clock very well? return them to OCUK?
Prime 95 for 12 hours plus and torture your best chip!!! I hope he did not waste his time with that Guff!!!
hello, 8pack, you think P95 for 12hr is nonsense? then how to determine the system is stable or not?
do you agree the 'batch theory' or 'PCB code theory'?
OcUK Staff
I use my system to prove stability. Prime actually only shows it can run prime. I would rather mine can game, code, unzip RAR files and do what guys really need stability for.
CPU batches yes for sure. All my 4770k that set records are all from two batches only for example. The same with my two 4930K one which does 6.3ghz XTU.
CPU batches yes for sure. All my 4770k that set records are all from two batches only for example. The same with my two 4930K one which does 6.3ghz XTU.
Even if I figure out the best batch no., how to make sure the one I'm about to buy is from that batch? Besides, individual chips from same batch perform quite different. I think batch theory is not feasible for end user. 8pack, does ocuk sell binned CPUs?maybe personally binned by you?
I use my PC for scientific computing.matrix, linear algebra, stuff like this. So p95 and linx perfectly model what I'm doing everyday. If system cannot survive p95 tests, I doubt it is truly stable. of course, one needs to monitor the system temp etc to make sure erroneous p95 results are not caused by overheat etc.
Yes but running the cpu against Linx, Occt, prime or IBT would give you a good level of stability when gaming. I used to stability test all of these at one time but now i just run 1hr of Linx on max memory. Thats enough for me to say my machine is stable in games that use the cpu heavily.
OcUK Staff
In need your incorrect. I have seen several CPU's running XTU, Linq everything fail in multi graphics. PCI X lanes are now controlled by CPU and these silicon torture tests do not show at all how GPU intensive stuff will work at a certain clock.
nniitt use your PC to run the tasks then as a stab test. The best way to show it is stable for your needs is run the modelling prog!!!
We do sell binned CPU yes in the 4.6 Bundle. This CPU is binned to exceed the 4.6 tag and most are doing 4.8 plus.
nniitt use your PC to run the tasks then as a stab test. The best way to show it is stable for your needs is run the modelling prog!!!
We do sell binned CPU yes in the 4.6 Bundle. This CPU is binned to exceed the 4.6 tag and most are doing 4.8 plus.
Im talking about cpu stability and not gpu stability. If this was the case then how come if i dont do stability testing then my system crashes with cpu intensive games?
As soon as i make my cpu stable in any of the stability testing software it works in cpu intensive games.
Im not saying that your wrong i just dont understand what you are saying?
Do you also sell binned chips without the bundle?
Associate
Aloha big man (8PACK) how goes the quest for even more world records 
any chance you could setup us guys waiting for the rampage black with some of your thought on what to tweak in this bios to get us descent OC'S on the 930k and 4820k pretty plz
Ineed they do not sell binned chips outwith the bundles, and what 8pack is saying most of the ones you use are synthetic meaning they only show they can run stable what they are programmed to run where as REAL world testing shows that its stable for what you need it to do
any chance you could setup us guys waiting for the rampage black with some of your thought on what to tweak in this bios to get us descent OC'S on the 930k and 4820k pretty plz Ineed they do not sell binned chips outwith the bundles, and what 8pack is saying most of the ones you use are synthetic meaning they only show they can run stable what they are programmed to run where as REAL world testing shows that its stable for what you need it to do
Last edited:
Hey 8 pack,
Is there anyway to get a decent clocking chip without part of a bundle (or playing cpu lottery)?
I've just ordered the 2 x 290x watercooled cards, and when this board is out will be grabbing it with a a 4930k and relevent waterblocks, but would be mighty disapointed to get one that wont clock well!
Soldato
Surely if they were to do binned 4930's they would likely to cost more than a 4960x so you may as we'll just grab one of those in the first place, although still not gauranteed i 'm led to believe there binned better.
in my case, a unstable system doesn't only refer to the one give me blue screen.
A unstable system may also yields erroneous results during calculation, I'm talking about a prog involving billion time of logic addition/multiplication. These prog take weeks even months to finish. p95 and linx have some mechanism to verify if the results are correct, in my case, I have no way to verify that. I cannot afford any mistakes resulting from the instability rather than flaws in the algorithm I designed.
Last edited: