is time for a new 4th gen chip/batch discussion thread...
Printable View
is time for a new 4th gen chip/batch discussion thread...
I think that we could list all reviews from yesterday:
Here is our review in Italian.
[*]http://www.techstation.it/hardware/a...swell-423.html
Could we do a list for each language?
Thanks
Alex
Made a trip over to MicroCenter (Westmont, IL) and picked up a 4770K/motherboard combo. They seemed to have all the same batch#... L310B492... this morning.
Maybe I'll get something setup/running later today.
Both of the ones I picked up there this morning were L310B492 also, asked if they had any CR but no luck.Quote:
Originally Posted by PolRoger
Malay or Costa ? (it's the same MC I'l go to )
Thanks
EDIT; NVM (Deux post)
Slightly OT but is anyone aware if Intel is using the same thermal adhesive to hold the IHS to the die with Haswell that they used on Ivy or are they going back to welds? If it is glue, I am looking forward to someone writing their experience de-lidding it and replacing the crap with a high quality TIM.
For example:
http://www.silentpcreview.com/forums...p?f=28&t=64782
Just picked up two L312B332 from Fry's will test tonight/tomorrow.
4770K L312B460
Max clock under water is 4.8GHz
4.9 and 5.0 not bootable into windows and stuck at logon screen :(
IMC up to 2933; 3000+ no post
http://img27.imageshack.us/img27/633...enchwprime.jpg
http://img22.imageshack.us/img22/1162/32mmaxmemaida.jpg
http://i458.photobucket.com/albums/q...l/DSC_0196.jpg
CPU: Intel Core i7 4770K
MB: Asus Maximus VI Extreme, BIOS 0607
DRAM: G.Skill TridentX 11-13-13-25-1T 2 X 4GB
GPU: Gigabyte HD7970, MSI's special BIOS, Driver Catalyst 13.6 beta
Storage: 2x Crucial M4 128GB raid0, FW 000F
ODD: LiteON DVD-RW
PSU: Corsair AX1200
Cooling: Custom watercooling - EK Supreme HF plate #6 CPU waterblock, 2x XSPC RX360 radiator, Swiftech MCP655 pump, EK FC7970 GPU waterblock, EK Multioption X2 Advanced reservoir, XSPC
1/2" ID 3/4" OD white tubing, 12x Scythe GT AP-15 1850rpm 120mm fan + Aerocool Shark 120mm fan, Coolaboratory Liquid Pro thermalpaste etc.
Casing: Dimastech Easy Dual V2 Bench Table
OS: Windows 7 Ultimate 64-bit SP1
Thanks owikh for this helpful post
My pleasure.Quote:
Originally Posted by XBrytanX
4770K @ 4.6 (undelidded) = 3770K @ 5.0 (delidded)
4770K @ 4.6GHz
http://img826.imageshack.us/img826/301/cinebencho.jpg
3770K @ 5.0GHz
http://i458.photobucket.com/albums/q.../Cinebench.jpg
http://i.imgur.com/KCMhNsS.jpg
2T
http://i.imgur.com/ZuSiyBh.png
1T
http://i.imgur.com/R0y4LvF.png
Still trying to get uncore to run at 5000
did you tried higher clocks with LN2?
Thanks but this makes my 3770k very sad...BTW the ram speed is higher on the 4770k but I don't think it made that much of a difference.Quote:
Originally Posted by owikh84
Testing atm with 5ghz, trying to get 5ghz uncore stable before full cold, whatever that is lolQuote:
Originally Posted by FlanK3r
It seems 5ghz uncore is going to require a lot of volts on this cpu, if it even happens.
Is it worth the upgrade from a 2600K?
MVIE flashed to 0006 bios, now 5.0GHz is bootable and benchable with 2 cores enabled. Max temp 72C.
4770K L312B460 @ 5.0GHz | G.Skill Trident 2800C11 2X4GB @ 2933 11-14-14-25-1T
http://img11.imageshack.us/img11/816/32m619881.jpg
My god seems a toasty one def not what i expected
3000 straight boot @ 1.71v
http://img812.imageshack.us/img812/2374/3000y.jpg
what is it for people who overclock their ivy bridge? The Haswell motherboard cost more, the cpu cost more, but after overclocking, you could hardly feel the difference, why would i want to spend money on it?
also the so call high end side is also horrible , the x79 has nothing for x77 users.
Looks like nothing for me for years. I am disappointed, i was ready to buy buy buy, now i will keep the money in the bank.
True StoryQuote:
Originally Posted by ma2ha3
OK... Here are a few screen shots from my 4770K sample...
47x ~6hrs "crunching" Rosetta 100% load:
http://i695.photobucket.com/albums/v...ps3819b1df.png
46x SuperPi 32M DDR3-2800:
http://i695.photobucket.com/albums/v...ps4aa830d5.png
47x Cinebench 11.5:
http://i695.photobucket.com/albums/v...ps61dc4804.png
nice pol if someone wants a comparison here my 3770K at 5GHz HT / 2666CL10 seems the 4770K at 4.7 pumps like the 3770K at 5GhzQuote:
Originally Posted by PolRoger
3770K 5GHz HT / 2666CL10-12-12-31 1T
http://i.imgur.com/ZbCjt88.jpg
still i love the 5Ghz number :)
Looks like L310 is a nice batch...if delidded the chip perhaps up more with nice temps
Ha! ha!...Haswell?
http://www.firegoon.com/u/tfmsq.jpg
I think there is quite a bit of chip to chip variation with HS... It is a little early to say but I've seen others complaining about the quality of the L310xxx batch(s)Quote:
Originally Posted by hersounds
I agree my friend, I have notepad here and writing info about batchs :), L310 and L312 seems good.
Although loves voltage, the L312B460 has a good IMC.
3000MHz 11-14-14-25-1T @ 1.84v passed 32m on air.
http://img62.imageshack.us/img62/2283/32m3k.jpg
3060 validated. That's the max of my chip's IMC could squeeze out.
http://i458.photobucket.com/albums/q...swell/3060.jpg
http://valid.canardpc.com/cache/banner/2825279.png
What kit are you using? CFR ic?Quote:
Originally Posted by owikh84
***Edit: I see it now... Trident X... CFR***
Yes, it's the $600 CFRQuote:
Originally Posted by PolRoger
http://i458.photobucket.com/albums/q...4/DSC_0087.jpg
2 modules at the bottom
My 4770K is proven to be a voltage lover.
4.6GHz @ 1.44v prime95 for 3 hours
http://img441.imageshack.us/img441/3046/p95463h.jpg
AIDA Stability for 3 hours
http://img51.imageshack.us/img51/1130/aidastress3h.jpg
So, it's time for a [WTS] :D
Delidded?
Nope, and I'm not going to delid any chip which can't do 5.0GHz @ lower than 1.5vQuote:
Originally Posted by WeaponX
Bath L307B, very bad (4600/4700), I have two :(, you know something about the L311B
4770 k Air cooled-PM(tomorrow will update batch number)-Batch L307B243
Kingston HyperX Predator 1866 2x8GB@2600MHz (never checked, i thing Hynix BFR)
Super Pi 1M
http://i.imgur.com/MQpfJRJl.jpg
32M
http://i.imgur.com/G66hueHl.jpg
Cinebench R11.5
http://i.imgur.com/bUDUyVil.jpg
seems good:), Im looking forward for batch info :)
I have a L311B424 4670k and it needs 1.325v llc Extreme for 4.5ghz. Pic below cpu has been delidded and liquid pro on core mx4 on block.Quote:
Originally Posted by zupernico
http://i1150.photobucket.com/albums/...dLiquidPro.png
Nice bro. So how much is the cooling improvement delidded vs undelidded?Quote:
Originally Posted by acebmxer
Btw, from where did you get the LinX? Mine is not working for Haswell the process is stopped suddenly for no reason.
Thanks
undelidded vs delidded was 10c drop in temp using Liquid pro on core and mx4 on block. Here is LinX - http://www.mediafire.com/?7w8xpkd233i85f8Quote:
Originally Posted by owikh84
Batch updated.Quote:
Originally Posted by lkozarov
Whats your temps (in load) Kozarov? Im still thinking, we have here L307Bxx in shps and I dont know If Il waiting or not...
Well hmm haswell adventure. so far tested my psc kit. it requires 0.02v more on m6e compared on m4e. 2133 C7. This is my fully stable clocks with aircooling maxing out at 95c on aida and 93c on prime. tested both for 2 hours.
http://img6.imageshack.us/img6/4017/haswell45ghz.png
same batch as owikh.
4.5ghz 1.24v.
so hot with this voltage? Uffff. What aircooling and ambients?
hmm its hard to explain but the heat u will only notice on coretemp. i guess because of the die and the tim issue. on casing side its actually cooler.
My temps:Quote:
Originally Posted by FlanK3r
4.5GHz@1.22V-Linx+AVX running without errors but hit 100C. Same conditions with Prime95 27.7 temps are under 80C.
4.7GHz@1.32V in Cinebench R11.5 is about 80-85C.
5GHz@1.4V 2C/2T in Pi32M temps are under 65C.
All temps are for hottest core.
That chip looks like a good candidate to IHS removal !
I removed IHS from two 3770K. On both temps dropped with 20-25C(I use them with liquid Pro without IHS). I gain ~100MHz with first and ~250MHz with second on air and Single Stage cooling, but on LN2 were much wicker compared with situation before procedure. One finished with broken die. So I prefer not to remove IHS for now.
L307B239 XXA1577 and XXA1529
Just need a little cold...
http://img23.imageshack.us/img23/341...eenshot007.png
^^ December air should help :D
Same here. :( I thought Sandy looked pretty nice compared to an i7-860 @ 4GHz but my upgradeitis bug had faded away a bit since the craziest hardware addiction years when I got into music/headphones so decided I will wait for Ivy. Then Ivy launches and I was mighty disappointed, now Haswell launches and it's the same story....Quote:
Originally Posted by ma2ha3
I had expected Haswell more to max out at like 4.8~5GHz on air for stable aircooling usage at maybe 1.3-1.35v in average, seems like now it fell a bit short from that, only special chips do that and you should expect more like 4.5-4.6GHz tops on air and anything more is a good bonus so basicly pretty much the same were Ivy maxes out typically on reasonable voltages/temperatures and then we are only talking like what, 5-8% IPC difference on average at the same clock, I was expecting a few extra % lead with OC advantages...
Now I feel like meh, could pick Sandy, Ivy or Haswell, not much difference. Years of waiting for nothing... oh well at least some of the Z87 mobos features Realtek ALC1150 chips with up to 600 ohm headphones support. yay :rolleyes:
Yeah I wanted a new cpu/mobo to play with and been waiting for Haswell, but I have a decent Ivy, benches 5 easy with 1.5v, and 4.7 prime stable with 1.3v. Just cant get excited to buy Haswell thinking I will likely end up with same or worse performance, and worse ocing. what a pisser... back to gaming forums until next release
so far all the vdimm that was required to be stable with my psc kit on p67 and z68 is up by 0.025v set on m6e.
still waiting for rma on my dmm.
the biggest benefit i saw was. lol unrar blu-ray and games now takes half the time from sandy 4.5ghz. 8gb in two minutes.
SS? Or colder...Quote:
Originally Posted by Dumo
SS -20c
4.5ghz upgrade difference.
http://i287.photobucket.com/albums/l...ps51278c1d.pnghttp://i287.photobucket.com/albums/l...psdd53c3af.png
L1 cache on steroids.
Quote:
Originally Posted by Dumo
Quote:
Originally Posted by tool_462
Those two look pretty nice Dumo. :up:Quote:
Originally Posted by Dumo
Makes me wish I had access to some better cooling.
Which is the batch please?Quote:
Originally Posted by Maxforces
L310b562
before
Sandy 4.5ghz 2133 C7 1.6v
http://i287.photobucket.com/albums/l...psdd53c3af.png
After
Haswell 4.5ghz 2133 C7 1.62v ( still cant bring this down)
http://i287.photobucket.com/albums/l...ps5f1e2ebe.png
made a mistake on the timings.. on before.
something aint right there...or AIDA bumped on new version 3.00 the mem bandwith...Quote:
Originally Posted by cstkl1
*edit yes its here http://www.xtremesystems.org/forums/...ghlight=aida64
its multithreaded.
http://i287.photobucket.com/albums/l...ps78fa1290.png
more determination now to get 2200 C7 fully stable. its only stable for benching but will fail prime in 12 minutes in all the time. Same thing happened before in p67 but the option to remedy that doesnt seem to be the same for haswell....
This physics score is with the i7-4770k at 4,5ghz?Quote:
Originally Posted by cstkl1
I think it is very low, i expected more.:(
I scored 12600 points with i7-3770k at 4,7ghz with 2133mhz ram.
Not bad score. Ivy and Haswell is very difference CPU.Quote:
Originally Posted by cstkl1
Testing on M6E all Malasia Chips. All Retails
Batch #
_______
L312B329 - tested on cascade at -115 idle @ -108c full load
2 pieces , one maxed out at 5756Mhz 1.78v. core 4C HT On ( No CB or CBB ) , the other one maxed out at 6005Mhz 1.82v.core 4c HT On ( No CB or CBB ) ( but i hurt the IMC and is wounded running Sammies at 2600+ cas 9 @ 2.05v mem and, cpu 1.82v almost dead , boots sometimes at stock settings , sometimes i get code 55 at stock settings no boot. Bad Batch lol , Lemon cpus.
L312B208 - 2 pieces , tested on Cascade and this batch have cold bug at -115c on cascade both cpus , it wont boot into windows , it would not enter bios ,it wont do nothing unless on stock cooling , not a good sign. To my friend L0ud Sil3nc3 who asked me earlier about cold bug , yes there are batches of cpus with cold bug my friend i have just find out, avoid L312B208 like a plague. We better wait for Costa Rica Chips :/ the only salvation and last hope ...
OK got 1.6v at 0.85v VSSA stable with the lowest timings etc. Awesome read/write/copy
http://i287.photobucket.com/albums/l...psf474829d.png
Still a few more timings to finish but lol after 6 hours kindda tired.
and because of the efficiency. 12k on physcis at 4.5ghz
http://i287.photobucket.com/albums/l...psd5b4d998.png
next up 2200 c7 and 2250 c7 the climb to 2400 is tiring but kindda worth it.
What would you guys say is max vcore for 24/7 on water?
The same question repeats every chip :)Quote:
Originally Posted by acebmxer
i always say that set your OC till u are pleased no matter the volts if you dont go over crazy temps of course, you can always achieve a nice OC without priming or linx to hell the chip, if you just want game stable well test chip with heavy games, most OC will collapse with BF3 for ex, you need to be stable to game some continious matches of it
Good luck !
Sergio
Got my CPU today 4770k Batch: L313B736
Still no ETA on when my board will be ariving
i tried 5.0GHz... and i made it :)
http://i.imgur.com/ecjTZZJ.jpg
didn't test it for stability tho :P and i don't think it would pass any test, but 5.0GHz for Haswell can't be bad ;)
wow,that is a strong cpu ! batch?
really sorry, but i have no idea :( i didn't checked before i've put it in and i bought it as OEM....
:eek:Quote:
Originally Posted by gravimag
:(
:(
Is there even such thing as OEM 4770K out there?Quote:
Originally Posted by gravimag
it is 4670K, not 4770k ;)Quote:
Originally Posted by jcniest5
WOw this VCore is looking goodQuote:
Originally Posted by gravimag
Yeahh 5Ghz huhu !!! CongratsQuote:
Originally Posted by gravimag
thanks :)Quote:
Originally Posted by andressergio
i wanna push it a little bit further... but i must change thermal compound under IHS ;)
careful i see ppl puting the CPU over a flat table and start cutting THATS NOT RIGHT you should grab CPU with your hands and cut corners and all there with your hands so its flexible and you dont hurt PCB, this is the best mode after 9 delid i didQuote:
Originally Posted by gravimag
What batch is it ?
mmm 5Ghz haswell not bad results :up:
Some run with MSI Mpower Max under WC:
SuperPI 32MB 5GHz 1.325Vcore 4c8t
http://img23.imageshack.us/img23/426...25ghz1325v.jpg
Wprimes 4.8GHz 1.235Vcore
http://img28.imageshack.us/img28/208...48ghz1223v.jpg
3dMark 06 4.8GHz 1.235Vcore
http://img824.imageshack.us/img824/5...48ghz1223v.jpg
BR
Nacho
give batch nacho! haha
and go coldddd
The batch number is L307B156, i try 5 of the same batch, and this is the only goodone Vivi.
guys, any idea what is the safe limit for CPU RING Voltage? i've just checked voltage measurement points and found out that with CPU ratio x46 and Uncore ratio x46/ 1.2Vcore CPU RING (auto) is at 1.344V!!! (default is 1.05V...)
CPU VRIN is at 1.744V so it didn't change much, but RING is just mad...
look at this madness over at hwbot's thread. i'm trying this batch tomorrow, our supplier has stock of this batch so im taking as many as i can
Quote:
Originally Posted by Alatar
congrats nacho that seems a great CPU !!!Quote:
Originally Posted by nacho_arroyo
Good luck Vivi :up:
Maybe in cinebench, the only benchmark where 4770k really shines.Quote:
Originally Posted by owikh84
maybe cauz its using the graphics on die ?Quote:
Originally Posted by LiquidReactor
I got batch L312B332 anyone have anything on this other than l0ud_sil3nc3Quote:
Originally Posted by Vivi
Any information for L312B334 ?
Quote:
Originally Posted by Phaseshift
From what I've seen, just about every L312B has been sorta meh. The only good chip I've seen so far has been from L0ud.Quote:
Originally Posted by Ziwro
RobQuote:
Originally Posted by Rob94hawk
with what u gonna cover the circuits on PCB near die to delid it ?
Nothing. Was going to do the vice method.Quote:
Originally Posted by andressergio
wich is ?Quote:
Originally Posted by Rob94hawk
http://www.youtube.com/watch?v=2jjxvY6Y-P4Quote:
Originally Posted by andressergio
Becareful with vttddr and vccin.
One proc dead. Two more to go, got a gaygay oc to test as need to check where the prob
is.
do you guys have speed step on?
Just did a quick super pi run on 32m
47x on 1.28v
will push for further when I get home from work.
using an h100i
i would never recomend that to anyone my friend and i also refeer to the circuits on the PCB u have to cover them to use liquid proQuote:
Originally Posted by Rob94hawk
I have NT-H1 and Liquid Pro on it's way. I may use NT-H1 on the die and LP on the lid to avoid any circuit issues.Quote:
Originally Posted by andressergio
As for the vice method IMO it seems less likely to damage the PCB than a razor.
not if you grab CPU on your hands most ppl put it over a flat surface and then push the razor and that how PCB is damaged...u need to grab always CPU on your hands so you curve it as needed i delided 9 chips with not a single mark on it lookQuote:
Originally Posted by Rob94hawk
http://www.xtremesystems.org/forums/...results/page29
Only @ 4.2 Ghz and core's hitting 80 on water! Yep. It needs to be delidded.
http://img.photobucket.com/albums/v2...psc40fb6a5.png