Where oh where has my BFR gone...

[Shiranui Gen-An]

Is it just me or has Hynix BFR completely disappeared from the market? It seems like most of the kits that used BFR are either discontinued or have switched to CFR. Just today I went to Micro Center to grab a couple of kits of Dominator GT 2x4GB 2000 9-10-9, all the pics online showed them as being Ver5.11 (=BFR) but what they had was 5.12 (=CFR). Seeing as how BFR is the only 2Gbit IC known to be able to do 1200MHz 9-11-10 or 9-11-9 it'd at least be nice to try to hit those timings if only for a CPU-Z validation Patriot 2000 C9 doesn't count, those are the weakest BFRs known to man

[Reefa_Madness]

Just today I went to Micro Center to grab a couple of kits of Dominator GT 2x4GB 2000 9-10-9, all the pics online showed them as being Ver5.11 (=CFR) but what they had was 5.12 (=CFR).

Are the Ver5.11 also CFR, or do you have a typo and meant to type BFR for Ver5.11?

Wonder if CFR is cheaper, so perhaps that's being used for everything but the very high end (2600/2666/2800) kits (with those getting the highly binned BFR, along with the Samsung ICs)?

[BeepBeep2]

They don't really clock much different...I think the good CFR can match BFR.

[Don_Dan]

I think it's a typo, v5.11 used to be Hynix BFR.

Hynix H5TQ2G83BFR is EOL and has been for a while afaik, those are all ICs currently in mass production:



The table can be found here.

[websmile]

Imho it is time that ddr4 comes because even best Hynix and samsung are simply getting on my nerves, when I read about CPUZ stable at 1200 C9-11-10 being good ics I know why I quit binning and testing...

[BeepBeep2]

Hypers can probably go CPU-Z "stable" at 1200 7-7-6

[Reefa_Madness]

Hynix H5TQ2G83BFR is EOL

Thanks for that info, wasn't aware that it was EOL. Would certainly explain why more and more CFR is showing up.

[websmile]

Hypers can probably go CPU-Z "stable" at 1200 7-7-6

I'm quite satified with mine running 32m 8-8-8 at this frequency - needs less voltage than Hynix for 9-11 cpuz

[Shiranui Gen-An]

Are the Ver5.11 also CFR, or do you have a typo and meant to type BFR for Ver5.11?

Yeah, I made a mistake there. Correcting now

[Shiranui Gen-An]

Imho it is time that ddr4 comes because even best Hynix and samsung are simply getting on my nerves, when I read about CPUZ stable at 1200 C9-11-10 being good ics I know why I quit binning and testing...

I'm sure the timings for DDR4 are going to "look" horrible until we see the bandwidth, the same as DDR3 did compared to DDR2, and DDR2 compared to DDR...

[websmile]

This will be the case, but at least on ddr4 we will be at the beginning of a new development and not at eol - as I am used to the horrible timings of the samsung/Hynix generation shock will not be that big

[Shiranui Gen-An]

Managed to find some very probably BFR, these sticks. Kitguru published a review of them over a year ago and they were obviously held back by the Sandy Bridge 1155 platform. I managed to push them to 1200MHz 11-11-11-28-1T without too much trouble (none of my CFR can do tRCD 11 at 1200MHz). I'm still hoping I can get at least a SuperPi 32M run with them at 9-11-10.

3820 35x128BCLK XPG 1200MHz 11-11-11-28-1T HyperPi32M.jpg3820 35x128BCLK XPG 1200MHz 11-11-11-28-1T AIDA64BW.JPG

Looks like these sticks have a hard tRP wall, they won't do 10 at 1200MHz. On the other hand they can run the rather Samsung-like timings of 10-11-11 at that speed.

3820 35x128BCLK XPG 1200MHz 10-11-11-28-1T HyperPi32M.jpg

[Shiranui Gen-An]

I couldn't pull the heat spreaders off but the flashlight test reveals these are actually CFR (AGAIN). Three solder joints per side. Still, tRCD 11 at 1200MHz is fairly strong for CFR.

[Shiranui Gen-An]

I finally managed to pry a heat spreader off one of these ADATA sticks. The aren't Hynix at all, they're Samsung HCH9. Explains why they do timings like 10-11-11.

[Shiranui Gen-An]

Well, I did manage to find some more BFR:

IMG_0201 (2).jpgIMG_0202 (2).jpg

Unfortunately, I should have paid heed to the OCaholic review. While I was able to push the sticks to 1200MHz on Ivy Bridge, the timings were loose (11-12-11).

[ZenEffect]

Hypers can probably go CPU-Z "stable" at 1200 7-7-6

want to see me boot 1250 7-7-6 on hypers?

[The Stilt]

Got myself 10pcs of Hynix OEM (bulk) 4GB sticks.
They are rated to PC3-10666 and have CFR chips on them.

All of the sticks do atleast DDR-2400 (12-13-11) with <1.5V, however some of them do not seem to like the voltage at all.
Some of the sticks do 11-13-11 with <1.55V and 11-12-11 with >1.6V. However raising the voltage above that or setting tRCD higher doesn´t make the scale any further.

A very recent production batch, manufactured in week 9/2012.

This is on K15.

Cannot complain really, paid 18 euros per stick