Lucky_n00b
-
Posts
852 -
Joined
-
Last visited
-
Days Won
11
Content Type
Profiles
Forums
Events
Blogs
Posts posted by Lucky_n00b
-
-
PM Sent
-
Reaaaaally wanted to go to World Tour either in Europe or Africa to see the overclockers there, hopefully in 1-2 years I can save enough budget to go. Good Luck for the event and keep pushing it
-
Hey guys,
I'm testing the 970 Pro Gaming for a SkatterBencher video. Of course I wanted to get some 5G action going. So far the Samsung kits seem to doing quite well. 333MHz 1:4 is straight boot and I'm now trying to get it more stable for 32M. I really want to try LN2 before leaving for the World Tour in South Africa :celebration:
How much volts for 5G Cinebench? Seems like nice FX chip & IMC
-
7870K are not bad, but of course, the high limit is few hudnred MHz lower. Example my 7870K can run 4700+ MHz stable and its good iprovement from 7850K (4440Mhz).
Interesting looks Excavator and Excavator+ ( (Bristol Ridge) as last Bulldozer chips.
3400 MHz Excavator core without tweaks od system etc:
Results is better than 5100 MHz Vishera with tweaks and better than Kaveri at 4500 MHz with tweaks in 32M
Of course, this is only in this benchamrk, so huge iporvement. Other will be less (+3% or so per clock)Based on that data, the X4 845 Excavator is going to be really interesting then
. I wonder when are those hitting the market, and how far can it be OCed by using BCLK (if it's possible). -
Great to see you getting lots of points with those setup Alby
As always, keep pushin' it!
-
I really hope that 40% instruction per clock doesn't come from the SMT (and they don't compare a 8c16T Zen vs 8c8t Vishera), and the clock scales well with cooling and voltage. Their 14nm capability is really unknown at this point.
The 28nm in Kaveri/Godavari still gives me nightmares (better IPC than previous gen but the CPU clocks scaling are not that good)
-
Definitely will be there
Anybody got early infos about what OC Events going to be there beside the WT Asia 2016 ?
-
No 6320 here, but I can bin the i3-6100 no problem
-
The problem is that there is no suspense except for a bunch of nerds interested in computers. You can sit back on your couch and casually watch football or gaming competitions, but overlocking ? Nah, forget about it.
A couple years back, people think that the only way to enjoy games are playing it on your own. And yet now we have trend of people watching other 'nerds' playing games
Also, many years ago games isn't considered as an 'eSports' like it is right now.Yes, the number of overclockers is really small, it's technical and sometimes difficult to explain to normal users. People nowadays wanted something easy and even they tend to have less 'motivation'to tinker with PCs (since using a PC is pretty mundane task for everybody right now).
But i'm thinking, if people can make a competition show about something mundane like cooking(like MasterChef), maybe if we put enough effort and consideration into it, we can make overclocking to be an exciting and informative shows ? No need to be a multi-million Dollar tournaments like DotA, but improving it an event at a time isn't a bad idea, right?
-
The only question is if we should add the value of prizes like hardware too.
While it may add some complications in calculating the prize, if it can be done I don't see why not, it's still a prize and sometimes we sell those prizes to buy our next-gen hardware
(I always keep the SSD though, can't have enough of those hahah ) -
Found this on my Facebook feed, interesting stuff. Data comes from: http://esportsearnings.com/.
- World of Warcraft: 470K for 78 players in 17 tournaments
- Tekken 7: 75K for 9 players in 2 tournaments
Wonder how they count the amount of players. Maybe they don't count the people who try to qualify for the tournaments?
Interesting, I just wonder if all the OC eSports tournament prizes on 2015 were to be totaled, how much would it be? Is there any way to calculate it Pieter?
- World of Warcraft: 470K for 78 players in 17 tournaments
-
The Gamer's Assembly Event in France seems nice
*calculating flight ticket to France*
-
Thanks for the kind words folks
Was a tough competition, but then again I believe things would've been different if Drweez got a slightly better IMC, XA got better GPU, Hazzan didn't lost his all his equipment in the luggage, or Splave showed up(hopefully Jack is better now, Allen).
So, this time I must've been lucky I guess
See you guys in another competition, keep pushing it
-
OPA another 120bclk good job bro!
Hahah it's the board doing its magic though
Hopefully we can see more awesome result from this board later from other overclockers, Cheap overclocking FTW!
-
Possible for David to provide Jagat with a test sample to verify @Lucky_n00b?
ALready did it a week ago, the board should arrive today
-
My sources tell me that this 'bclk unlocking' should be a legit feature from SuperMicro C7H170-M motherboard, if I remember it right it's called SuperMicro SuperJet Technology or something.
Really interested in how they did it though, kudos to SuperMicro to make something like this! :ws:
-
So I've gotten myself 3 Fury Xs and a regular Fury. Since Extreme Addict said he didn't get any cold bugs I figured I would try at least 1 of my cards on LN2.
Now because I'm kinda short on money(I just spent it all on Fury Xs) I was wondering how much difference there is between a slim pot and fat pot if I'm just trying to run the card as cold as possible.
From what I know a slim pot would be better for keeping the card really cold and a fat pot is better for keep the temperature steady.
I don't bench 3D much, but here's my experience with the pots:
Slim pot are okay-ish when you wanted to do low-TDP GPUs like GTX 750/Ti/GTX 960. Slim pot usually use much less ln2 when pulling down the temps, you also get good temp response with these 'small' GPUs.
But if you use slim pots for heavy gpu (980Ti/ 290X), it will be hard to control during load, and temps will jump all over the place, even if you 'torched/glazed' the pot.
Fat pot is the way to go if you wanted to do heavy GPU, it uses lots of ln2 when pulling down, but the fine control it provides during full load will be worth it
Also it might be better to 'invest', buy good pot from the start, it will last quite a while.
Hope it helps,
-
wow, awesome score, really efficient
Hey thanks man
Will try harder with better CPU and AFR next
-
Very Impressive
:eek:Must have taken you days if not weeks to get all that info.
I work as a reviewer, and memory review has been something I've done a lot to slowly increase my skill as an overclocker. Despite doing this on daily basis, I can certainly say the information you put in the first post is WAY much better than what I know right now.
Amazing guide man, keep pushing it!
-
Hello,
I'm searching for Intel Skulltrail platform which consist of:
- Intel D5400XS Motherboard
- CPUs, can be the 2x QX9775, or a cheaper 2x Xeon E5410 or equal would be preferred
- 2 or 4 sticks of FBDIMM rated at least DDR2-667 speed, but a DDR2-800 Kingston kit would be nice.
If you have one and wanted to sell it, please PM your offers include express tracked shipping to Jakarta, Indonesia.
Best Regards,
Alva
-
Oh, and one more just to make sure: All components have to be retail and not ES? Or ES allowed?
-
Question:
Stage 2 - Fire Strike IGP: use 5 different integrated GPUs, each using a different CPU socket. The ranking is based on the average of the five results. There are 40 points for the winning country
FM2 and FM2+ counted as different socket?
Stage 3 - Geekbench3 Single Core: use 5 different memory types in the Geekbench3 single core benchmark. The ranking is based on the average score. There are 40 points for the winning country
DDR3L & DDR3 counted as different types?
Stage 5 - GPUPI 32B Full Out: this stage requires 1 score and allows full out hardware selection. There are 30 points for the winning country
Server/workstation components allowed?
Thanks
-
Toolius, Toolius everywhere on the top benchmark scores of the day! Keep up the good work man
-
Just trying to give suggestions here, maybe next time HWBOT(and its partners) need to discuss the definition of 'retail availability', and add it to the rules. IMHO, the board can be considered 'retail' if at least these two criteria were met
- The manufacturer says so, and already list the final product detail on their website, AND
- There's at least 2-3 online stores that ship worldwide in each region(USA,EMEA,APAC) who already sold that product for at least 3 weeks.
And would be better, if the vendor can give help to the overclocker who wanted to buy that specific product that will be used for competition
Excavator
in Steamroller / Kabini / Excavator OC
Posted
@Massman
Just got back from the store binning some g4400 for Pro cup 32M stage, and found one X4 845 there
Fastest in AMD 32M 5Ghz before Flanker come back with his better score hahah:
http://hwbot.org/submission/3151415_lucky_n00b_superpi___32m_athlon_x4_845_11min_20sec_358ms
I don't expect much overclockability from this design, IIRC AMD mentioned that CZ's design will sacrifice frequencies to made it more efficient in the lower TDP range. This one has it's turbo rated at 3.8Ghz 1.475V VID, and mine already need 1.5V to pass 105Mhz BCLK (4Ghz-ish), so the default 3.8Ghz could be it's optimal frequency range vs volt vs TDP already.
Expect more test next week, need to write some other articles first