Okay, I'm curious now. What are the fastest MFR based SuperPI 32M runs out there?

Not sure what you mean,highest MHZ or highest efficiency?Cause single-sided memory scores so low it's not worth the effort.8 GB double-sided MFR high-binned should be interesting to see how they perform...

Full out => best score

4G / 5G => best score, therefore best efficiency

 

I'm curious about these SuperPI 32M runs because there seems to be a battle going on for the best MFR based SuperPI 32M runs.

Dumo is got MFR pretty dialed at 5Ghz, I want to say he was at 6:06:xx or lower off the top of my head.

5 GHz by l0ud_sil3nc3:

 

 

4 GHz by Splave:

 

 

Those are the only runs with MFR in the XS 4 and 5 GHz 32M Low-Clock-Challenge.

tried it but never 4g and 5g

also nick tried some C8 i think

Waste of time,nothing can match ln2 cooled 2600+ C6 psc and bbse.Maybe samsung 3000 9-12-12 in theory because practically highest benchable freq i saw was splave 2933

Full out => best score

4G / 5G => best score, therefore best efficiency

 

I'm curious about these SuperPI 32M runs because there seems to be a battle going on for the best MFR based SuperPI 32M runs.

 

i think its more just superpi to test out staiblity of MFR overclocks, lol

 

I could be wrong tho

Edited July 8, 201311 yr by sin0822

Thanks for the info. So about 15 seconds slower at 5GHz.

 

Well, that settles it then. I'm not going to be running any MFR for SPI performance runs

MFR is the AMD of the memory sticks. High clock - no performance.

posted this morning by Dumo, and this even has nV control panel in the tray

 

Edited July 9, 201311 yr by TaPaKaH

Thanks for the info. So about 15 seconds slower at 5GHz.

 

Nah, it's more around 5 - 10 seconds I guess. L0ud_sil3nc3's run was done on B0 CPU and with lower NB, fully tweaked on C0 with NB 1:1 should be noticeably faster.

 

Another 5 GHz run by Dumo:

 

 

I'd like to see a fully tweaked 1600MHz CL8 run, should be quite interesting.

MFR is the AMD of the memory sticks. High clock - no performance.

 

Bingo. Typing up a review on the 2933 G.Skill kit. Results are disappointing. They are designed for high frequency and nothing else. They fail to do the one thing a more expensive, higher-rated kit should do - increase performance. However, for people that enjoy memory clocking, they're a heck of a lot of fun.

 

So it depends on your goal. If you want better benchmark times, look elsewhere. If you want zomg memory frequency, go MFR.

Price performance figures must be aweful. For some high-clock 4x4GB high density kits you're paying well over €1000, with not a single bit more performance. In fact, more money brings lower performance.

 

I have a strong urge to fill the "rant about reviews"-thread with a list of all sites that recommend these MFR based memory kits

If you compare them to single-sided Samsung or double-reject CFR for daily use, MFR is not too bad actually

If I was running a GPU setup with 12GB+ of graphics RAM, 4x4GB or more of capable system RAM has a place.....but ermagherd.... not at 1000 euros. Not even $1000.

 

What about people NOT running their memory under LN2? How do xFR sticks shape up against PSC & BBSE? The story changes, right?

MFR is the AMD of the memory sticks. High clock - no performance.

 

and siggggged

 

 

Will try 4x4 soon

Gotta run it @ 3150+ tights to get subs 6s 5Ghz

any of you benched 8 gb mfr modules?Granted they won;t reach single-sided speeds,but there must be some good modules in the wild ...

?

any of you benched 8 gb mfr modules?Granted they won;t reach single-sided speeds,but there must be some good modules in the wild ...

?

 

I've been wondering the same thing myself... Are these all of these MFR chips going so far because they're only single-sided?

The ones to look for are mfr pbc 22x~25x single sided which mostly can boot @ 1.85V+

Here you go guys: http://www.overclockers.com/gskill-tridentx-ddr3-2933-memory-kit-review . Oh, and Massman, it received the 'no award' award.

Read it already!

 

The conclusion sums it all up quite accurately. For pure MHz it's a go-go, but if you want performance, you can go 40% cheaper with better performance. Nicely done! :celebration:

I feel like I just won the special Olympics of 32m benching:p: