CFR scale from cold way better than Samsung, so no wonder these do 1450 CL9 - 1400 11-13-13 is the highest CFR bin ever.
I think there is even a CFR LN2 profile for 1400 CL8 in Maximus VI-series BIOSes.
Printable View
CFR scale from cold way better than Samsung, so no wonder these do 1450 CL9 - 1400 11-13-13 is the highest CFR bin ever.
I think there is even a CFR LN2 profile for 1400 CL8 in Maximus VI-series BIOSes.
Remember kintaro run on 2800 11-13-13 gskill trident kit with HYNIX BFR
http://www.xtremesystems.org/forums/...fr#post5152981
full pot with hynix BFR did 2720 8-11-9...But for some reason,there are samsungs out there that can do trdrd4 at 2800+ when cold,so you may never know unless naked..
I just tested my CFR - these do 32M at 1400 9-13-11 on air, so 1450 on LN2 looks very realistic.
I came across this kit in a f/s thread at OCF and was wondering about the ic for "9200" Powerchip?
http://i695.photobucket.com/albums/v...0.jpg~original
Should be powerchips, the first digit can vary, even on ddr3, as this is ddr2 9200 should indicate PSC, which were used on DDR2 2GB kits by GSkill frequently
Samsung?
http://i.imgur.com/INci1nk.jpg
Yes. I had two of these quadkits, were nothing spectacular.
Stripped 2x2GB kit of G.SKill PK Series. They use exact the same Powerchip IC as DDR3. They're marked 1030 (DDR3 are 1040), I saw other kit (PQ Series 1000MHz) marked 0630, so it means to me, that DDR2 has DDR3 markings minus ten, if they're based on the same IC. Tomorrow I should receive a board, so I'll do small OC test :D
http://imageshack.us/a/img842/9249/5r4i.jpg
http://img547.imageshack.us/img547/9430/xx84.jpg
Hey... I've seen those before!
:D
Just RMA'd a kit of 2200C7 Pi sticks.
recieved back a new kit, green PCB, 8117
13331200xxxxxx
still PSC from quick boot timings (2600 8-12-8)
F3-16000CL7D-4GBPIS
What about this kit, BBSE/PSC?
Specs are 1000 MHz 7-10-7-25, right? Pretty sure they're PSC.
No, sry should have told that. The seller claims 7-9-7-24. Typical BBSE timings, but I hope for PSC :)
These ones http://www2.ncix.com/products/?sku=53572
1000 MHz 6-9-6-24 is a typical PSC bin, 7-9-7-24 can be PSC as well.
Let's hope for that. Will probably buy them together with 4x2GB G.SKill ECO 1600-CL7-8-7-24
G.Skill Flare 2000 CL7 has the same timings and it's PSC. I got it, s/n is xxxx1040Quote:
Originally Posted by Calathea
Trident X, 4x32GB 1866 C8-9-9-24 1.6V. Samsung HYKO. Sept 2013 build/batch.
The heatspreader fell off, so that's how I know hehe.
Picture:
Attachment 131354
I hope Samsung comes out with a better 4Gbit IC to challenge Hynix MFR, or at least that can clock like their 2GBit D-die stuff.
1500 or 2500 or something else as digits 5 to 8?Quote:
Originally Posted by Cookiesowns
Since HYK0 it is a DDR3-1600 part, I would expect the "prefix" / 5th digit to be a "2", therefore I would expect to see a 2500 in the 5th thru 8th spot of the S/N.Quote:
Originally Posted by Don_Dan
I don't think that whether the part is a 2Gbit (used in double-sided 4GB modules) or 4Gbit (used in double-sided 8GB modules) matters on the GSkill coding scheme, does it? TaPaKaH posted a while back over on OCN about the "prefix" or 5th digit of "1" being for the DDR3-1333 parts like Samsung HCH9 or Hynix BFR and the "prefix" of "2" being for the DDR3-1600 parts like Samsung HYK0 or Hynix CFR and so far that seems to hold up, whether it be on 4GB or 8GB modules (single, or double-sided).
Reefa 100% correct .
The following is a copy of a post that I made over on OCN shortly after Sam made his comment about the distinction between the "1" and "2" prefix in the IC code part of the S/N.
Could you guys check it out and critique it, or if you have additional info to add, please don't be shy.
Original post:Quote:
Following that logic then, a 2x8GB kit of Hynix based MFR PBC would have a S/N with a "2400" in it, but one made using the older MFR H9C (if used) would sport a "1400". Would you agree?
http://www.hynix.com/products/comput...=computingDDR3
Over at XS there was a pic posted of the ICs of one of these 2666 C11 kits with the "1400" and it was in fact CFR H9C (a 1333 part if I'm reading the info from the Hynix site correctly).
http://www.xtremesystems.org/forums/...=1#post5153644
Therefore a 2x4 double sided kit with Hynix timings and sporting the "2400" number should be expected to be using CFR PBC...a 1600 bin part?
http://www.hynix.com/products/comput...=computingDDR3
So if you know whether it is a single-sided, or a double-sided module, then based on whether you are looking at a 2x4GB kit, or a 2x8GB kit and whether it has the 1400/2400 for Hynix or the 1500/2500 for Samsung, you have a pretty good chance of guessing what is under the hood.
Does this look right?
For 4GB modules:
single-sided 1400 = MFR H9C (1333 part)
double-sided 1400 = CFR H9C (1333 part)
double-sided 2400 = CFR PBC (1600 part)
single-sided 1500 = the Samsung 4Gbit HCH9 (1333 part)
double-sided 1500 = the Samsung 2Gbit HCH9 (1333 part)
double-sided 2500 = HYK0 (1600 part)
for 8GB modules:
double-sided 1400 = MFR H9C (1333 part)
double-sided 2400 = MFR PBC (1600 part)
double-sided 1500 = the Samsung 4Gbit HCH9 (1333 part)
double-sided 2500 = ??? (I don't know it)
There is also a Hynix AFR that I've gotten on some generic 8GB modules and it is like MFR so it is possible to get these instead of MFR.
http://www.hynix.com/products/comput...=computingDDR3
There are also some Samsung variants to HCH9 that have popped up from time to time, so there are no guarantees, but generally speaking, most Hynix based stufft these days seems to be either CFR or MFR, and most Samsung would be either HCH9 (2Gbit or 4Gbit) or HYK0 (Edit: again, 2Gbit or 4Gbit).
Edit: If any of you guys have GSkill kits that you have the heat spreaders removed, could you check them against what I have above and confirm Sam's statement.
http://www.overclock.net/t/1268061/o...#post_20158789
Oooh yeh, these ones have potential :)Quote:
Originally Posted by Calathea
http://i1050.photobucket.com/albums/...ps9c697d02.png
great !!!Quote:
Originally Posted by Calathea
Thanks. I see now that the SPD have the specs that don dan guessed, confused.. whatever, they perform good :D The ECO's didn't seem so good, around 2400 8-11-8 1.8 volt but maybe I can push some more.
i tried the echos and most i get is 2450Cl8Quote:
Originally Posted by Calathea
I can't see that happening anytime soon, is see Hynix is using mfr for ddr4 http://www.hynix.com/products/comput...=computingDDR4, also if it's not my imagination I think I saw mfr on my hd 6950 which is gddr5. So therefore I have a slight suspicion that mfr has a lot more potential.Quote:
Originally Posted by Shiranui Gen-An
http://abload.de/thumb/122212067-11-7-28166vjvpf7.jpg
12221200 Pi 2000C6 kit I received from rma a while ago, never bothered testing these fully untile a week ago, 1200 7-11-7-28 auto 1,65v 32M, trcd10 at 1200 boot ans 1M, 1300 8-12-8 TWCL6 1,84v on Extreme. Think I will sell these, most of my mems are just catching dust
Nice websmile, high quality air clockers :)
today's lolwut-moment courtesy of G.Skill:
F3-17600CL9D-4GBXMD rated 1100MHz tRCD 9 with PSC (1103104Lxxxxxx)
http://abload.de/img/picture00457unw.jpg
I've never seen mem like this. Nice!
Envoy? depuis mon GT-I9100 avec Tapatalk
It gets better, ICs have stripes on sides and pulling one of the heatspreaders I was able to see an E904.. marking, but overclocking seems much closer to PSC.
G.Skill F3-12800CL7T-6GBPI with 0640 in the serial. Was hoping for BBSE, but these are clearly BDBG.
http://i.imgur.com/kr9yLh3.jpg
http://i.imgur.com/lsRSWy1.jpg
Make them fly, some of these are not too bad :)
This is the first kit to pass the PSC 2400 1.85V preset from the OC Formula! Oddly enough, 1.8v+ wasn't stable enough to pass Super Pi 32M, I actually needed to *lower* the voltage.
https://5p7fug.dm2301.livefilestore....SuperPi32M.jpg
Sent from my iPad using Tapatalk
^ That screenshot is incredibly small... :D
I think it's a PI kit, but which one exactly?
12800 CL7
Just bought some ripjaws 2133 9-10-9 1.5V kit,clearly with psc 1040.Hoping for a good set,fingers crossed :)
Thanks! :)
That's what I thought but I wasn't sure.
The 2133 9-10-9 1.5V kit should be quite decent, they're PSC for sure. :up:
Yes sir,by the week it\s XFG ic's :)
Bought these in this week:
Attachment 131890
2x2GB RipjawsX 2200Mhz with original box and fan. I can't wait to test them :) I'm wondering if the can beat my best PSC kit - Axeram 2400MHz, which do 2600 8-12-7-27 twcl6 on 1.83v. So far I've never seen anything better.
this kit certainlyQuote:
Originally Posted by bartx
based on PSC t-series chips. Anyway will show us your results )
How you figured it? I bought these, because they have 1040 marking and it seems to be nicely binned PSC - that's my opinion.Quote:
Originally Posted by MaJ0r
I don't know exactly. I always ask Sam if I want to know what chips have some memory )Quote:
Originally Posted by bartx
I'll be happy if I make a mistake because I can buy the same memory in Ukraine
Mediocre psc,you have to figure out that gskill already had pi 2200 7-10-10 kits,also higher binned kits like 2133 7-10-7,2000 6-9-6,2300 8-11-8,2400 9-11-9,2400 8-11-8 ...I notsure if t or x series,for sure you need to test them out...
Have you tested your 1600 CL6 PI kit yet?Quote:
Originally Posted by Shiranui Gen-An
Quote:
Originally Posted by bartx
This kit can be either PSC Xxx or PSC Txx. Even if they're based on PSC Xxx they aren't a great bin. You have to keep in mind that a red heatspreader always means low end on G.Skill RipjawsX sticks.Quote:
Originally Posted by MaJ0r
Yeah, G.Skill was crazy about binning PSC Xxx ( mostly 1.65V ): 2500 CL9, 2400 CL8, 2400 CL9, 2300 CL8, 2200 CL7, 2200 CL9 1.6V, 2133 CL7, 2133 CL9 1.5V, 2000 CL6, 2000 CL7, 2000 CL9 ( @ 1.65V or 1.6V ), 1866 CL8, 1800 CL9, 1600 CL6, 1600 CL7 1.35V, 1600 CL7, 1600 CL9, 1333 CL7 1.5V... I'm sure I missed some. :rofl:Quote:
Originally Posted by Alex-Ro
Unfortunately some of these specs can also be done by PSC Txx, for example 2200 CL9 1.6V, 1600 CL6 or 1600 CL7.
Booh, Dan :p: - you missed 1800 C7 and 1866 C9 :D
Yeah, I actually have 6 sticks of the 1600C6 Pi and now three Trident 2000C9. Testing them all one by one, all will boot 2400 7-11-7 but need 1.95V and only two of the Tridents were stable enough to load Windows and attempt Super Pi 32M, but failed after loop 1 on Asus M6E.Quote:
Originally Posted by Don_Dan
PSC X-Series, mine are 8155 PCB and very nice handle high voltage as well :)Quote:
Originally Posted by bartx
Damn it, I missed the most important ones... :DQuote:
Originally Posted by websmile
Thanks for the info! :)Quote:
Originally Posted by Shiranui Gen-An
When were yours made?Quote:
Originally Posted by Splave
12031040 is confirmed PSC THY532A3G-A.
Interesting. Do you have any screenshot of them?Quote:
Originally Posted by Splave
nvm mine are blue
yeap )))Quote:
Originally Posted by Splave
Got some Ripjaws 2000 C9 in today. Serial has 0640, BBSE as expected.
http://i.imgur.com/1R2E5gk.jpg
http://i.imgur.com/nL6Iy6n.jpg
Still can't do anything like 2600 8-12-8 or 9-12-9 :(
On which platform?
Asus Maximus VI Extreme + 4770KQuote:
Originally Posted by websmile
You can always try on second platform like ivy-e or 1155 if you have rig at hand, but it is very likely that these are simply not too good bbse that fail :(
Yeah, I get code 55 trying to boot 2400 8-11-8 up to 1.85v and the board doesn't seem to like to train if you go above that on voltage.
It turned out to be a crap - Powerchip T series for sure, they don't even boot at 2400MHz 8-11-8... :cool:Quote:
Originally Posted by bartx
Could be Imc. I have 2666 8-11-7 blue ripjaws. A couple of my chips can't run trcd 11 other have no issues (4770ks)Quote:
Originally Posted by Shiranui Gen-An
The IMC would be able to do tRCD 11 on some ICs but not others? That's very strange.
This is Haswell :) - the big progress :rofl: - compability is issue at newer systems, because of this double checking is always good idea if possible, as mentioned above
I think much better set up good Sams D kit into rig and test it. If memory can 2600 MHz + it obviously not IMC issueQuote:
Originally Posted by websmile
F3-12800CL6T-6GBPI X-Series PSC on Asrock Z87 OC Formula, PSC tight 2400 1.85v profile.
http://i.imgur.com/gThGLkf.jpg
Dear fellas,
I have in my possession a F3-2400C9D-8GTXD (TridentX 2400 CL9-11-11-31)
Manufactured 2012 Dec. Serial number is 1252250P39xxxx (Country France (if it could affect the serial number))
It's close to the 2500 Samsung HYKO digits but not. I still wonder if it's Sammy or Hynix under the hood and i don't want to remove the heatspreader.
I have a 4670K tested IMC 3000 OK with some TeamGroup MFR SS
So far with this TridentX kit and 1,65v :
2600 CL10-13-12 1T ; SPI 32M OK
2666 CL10-13-13 1T ; SPI 32M OK
2800 CL11..........1T ; SPI 1M OK and fail to pass 32M
Any idea what IC's could it be ? If Sammy "D", what could be the 24/7 safe voltage?
Thanks.
It is not unusual, for what reason ever, that GSkill sticks sold in france and iirc in some other countries like italy for example have a letter in part of sn that indicates ics, I have seen this several times before on bbse and psc kits for example. Results and serial number 250P indicates Samsung D2gbit, Hyk0. These can take a lot of voltage for 24/7, but on the indicated 1,65v you can do nothing wrong. Max voltage for ics in Samsung whitepapaer is 1,975v iirc ^^ - because of safety reasons, IMC etc I personally would not use more than 1,75v for 24/7, but 1,65v is within warranty and should be enough for good result
Thanks websmile ;)
I'm a bit disapointed with my kit, i've seen better HYKO's here and there. At least, they perform a bit better than my ex MFR.
A part in me is telling to give them more juice for H24 ^^
IMC, IMC,.. what a dilemma.
I tought the juice for the IMC is from VCCSA, System Agent or whatever it is called by the board manufacturer. High VDIMM have an impact on IMC life ?
To be honest, this is hard to predict, just discussed this with a friend, normally Ivy IMC can take lots of vdimm without degrading, tests and benches above 2v mem voltage on air and it still lives. I don?t use high voltage on 24/7 so I can only tell from benches and tests, if you decide to use high voltage for 24/7 like 1,85v++ on your mems, be aware that there might be a risk and let us know if you have any problems in future :D.
Perfect Storm 2133C9 Elpida BBSE
http://i.imgur.com/ygh0TQi.jpg
http://i.imgur.com/ySdTDYn.jpg
Is that the kit that you linked several days ago...the one that you were threatening to buy? :)
http://www.ebay.com/itm/G-Skill-Perf...vip=true&rt=nc
Guilty as charged :p: Sadly though, they can't compare to the 2133C8 kits from my initial testing. BBSE is a strange IC.Quote:
Originally Posted by Reefa_Madness
Hope you are more lucky than I was years ago, my two triple kits failed 2133 7-9-7 on sb
2133C8 seems to be an exeption. Easy on air, crapped out with cold.Quote:
Originally Posted by Shiranui Gen-An
I don't know why this was so hard to do on Hasfail.
http://i.imgur.com/p4dT6VI.jpg
The DDR IC guide was so right, BBSE doesn't like going much over 1.80v. I miss the old days when more juice solved everything :(
Quote:
Originally Posted by Shiranui Gen-An
You bought them! :)Quote:
Originally Posted by Shiranui Gen-An
At least they're beautiful... ^^
The lower-binned BDBG 1600C7 Pi kit can match or surpass them:
http://i.imgur.com/11JUaPx.jpg
Wow! That's surprising, I expected the 2133C9 PS kit to be a bit better. Have you tested 2133 7-9-7 on the PS kit?Quote:
Originally Posted by Shiranui Gen-An
2200 7-9-7 gave code 55 on Ivy Bridge, and I don't think Haswell can run those timings at that speed.
These Perfect Storm sticks are confusing as heck. They failed Sam's binning method on Ivy Bridge, but they pass the M6E BBSE 2200 preset pretty easily, and I can boot with the tight 2400 1.85v preset, but can't get it Super Pi 32M stable. 1M is easy though.
http://i.imgur.com/IWOq1bB.jpg
http://i.imgur.com/9m7CVpv.jpg
They also don't mind booting on high volts (1.85v+), so it seems like they can be good on air...I wonder if I just need to tweak something.
Have you tested more CPU input voltage?
My Giga's & Asrock like 1.9+ I find you can be unstable then bump input up a notch gaining stability untill you got too much voltage & then you start to become less stable again.
Anyone have an idea about these G.Skill Tridents? 2000MHz C9's
0640 is unnamed in the first post other than Elpida
9380640148809
9380640148810
10330640134391
10330640134392
BBSE and not good ones, you should be very lucky to find a good match of 2000C9 BBSEs
Figured as much, thank you.
A friend of mine received from RMA a f3-12800cl8d-gbrm kit (2x2GB rated 1600 8-8-8-24 1.60V), because the previous one (BxSE) ended up having a consistent faulty address (at any settings) that hci memtest wouldn't always pick up unlike memtest86 under DOS due to full cover.
Now, this serial starts with 13431200 (2013 Oct on hologram) which seems to be PSC on the list (x200?), but these sticks are single sided, with 8 ICs each.
They seem really bad compared to the previous kit. I don't know what ICs are under the spreaders. My testing was very limited as there were meant for a stock 955C2 under a M4A89GTD PRO (this same C2 could run 2x4GB 1600 7-7-7 D9Qxx without a problem at 1.50V).
1600 8-8-8-24 1T 1.60V - the rated timings are fine
1600 CL7 always fails, can't even POST at 7-7-7, 7-8-7 or 7-8-8 2T with 1.60 or 1.65V.
1333 CL6 6-6-6-24 1.65V fails to boot windows, and craps out in memtest86 immediately.
0640 is unnamed because it can mean different ICs on different kits, but hiwa is right of course, 2000 C9 with 0640 are BBSE.Quote:
Originally Posted by Gunslinger
There are some PSC 2Gb ICs, 1333 7-7-7-21 @ 1.50V would probably work as well.Quote:
Originally Posted by AlleyViper
Yes, higher density (2Gbit/4Gbit) PSC are no good clockers, much like all Elpida 2Gbit/4Gbit ICs on which these are most likely based on.
@Don_Dan As the rated 1600 8-8-8-24 1.60V does fine, 1333 CL7 seems less interesting, that's why I didn't bother to test other than CL6.
@TaPaKaH I presume you mean the Elpida BFxx I've seen on some KVRs, but there isn't much testing around.
There are also ICs like Elpida BCSE, these are older and did not too bad, I had these on 1600 7-7-7 rated 2x2gb single sided kits and they did around 2000 8-9-7 and up to 2400 10-10-10. But I assume your new psc do not have this quality, and I am not sure which Elpida IC might have been their inspiration
@websmile: They seem inferior to those BCSE by far up to 1600, and unfortunately if they scale well on MHZ with added latency, this isn't the setup to test them.
Just got my hands on a F3-12800CL7D-4GBRM ripjaws blue kit (7-8-7-24 1.60V 10491040..., probably PSC T?). Finally, I guess these will replace my OCZs, they do 1600 6-8-6-24 1T 1.55-1.60V under hci memtest without any tweaking. Stock CL7 (@1T) 1.40-1.45V seems solid, pretty much like Eco on AM3.
http://abload.de/thumb/jd8012993ps3n.jpg
Some of the last 1040 GSkill psc, used ICs are XEJ
some XEJ are very nice under cold but no 1400 C6 here :S
I figured out the issue with the Perfect Storm sticks. With auto subs, if you set 7-9-7 at 2200, the Asus boards try to set tWCL to 7, which is actually ideal but too tight for most BBSE from what I've read, hence the code 55. If I manually set tWCL to 8 (which is what the Tridents default to) they'll boot just fine.
Your mems are better than mine, one kit I had failed also on profiles and handsettings, cas and trp lowering needed that much voltage i had to run them 10-11-10 at 2500 on ivy...
P.S. What was indeed funny is that these were not bad mems apart from that, I could run them at 1000 8-8-8 for example and they had quite good trcd limit - might have been good mems for cold, which I do not use
Battled with this exact issue this weekend with my 2133 9-9-9 Tridents. Run without issue and tight subs at 2200 7-9-7 and lowish (1.715~) volts but as soon as tWCL is set to 7, nothing can get them to boot.Quote:
Originally Posted by Shiranui Gen-An
I think BBSE need cold and crazy volts for tWCL 6~7 with CL7.
hi
i will receive this kit from rma, can someone let me know what are they and should i keep it ?
F3:12800CL6D:4GBXH
S/N: 140218000070204
140218000070203
thanks for any help
Elpida according to the sn, could be BBSE or a variation of other ics, if they are worth keeping them only tests will show
hi websmile
thanks for your time
i have seen in the first post that 1800 was for elpida, but was hopping someone knew something more about them : -)
thanks for your help
I'm going to purchase a quad kit used memory and they have this serial, it should be HCH9? Does this sticks can do 2600+ or due to high density shouldn't except much?
http://i.imgur.com/uRPCdRR.jpg
These should be Samsung 4Gbit B-rev "HCH9". You shouldn't expect more than 1200-1250MHz out of these.
I finally managed 1300MHz 8-11-7-28-1T out of some of my BBSE, the Perfect Storm 2133C9 to be exact. I used some of the IOD/IOA/SA volts from the Z87 OC Formula thread and found settings that worked, and the board trained with high VDIMM no problem this time.
http://i947.photobucket.com/albums/a...ps31246b86.jpg
Now if I could only find a way to get some of my PSC to do 2666 8-12-8-28-1T and pass :(