Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.

paulrbeers

macrumors 68040
Original poster
Dec 17, 2009
3,963
123
Very interesting benchmarks. I have screen prints, but I don't have time to upload them tonight so you'll just have to take my word for it...

Geekbench 32bit (I don't own a license):
Stock 4GB @ 1600mhz - 10734
16GB @ 1333mhz - 10958 (tried it again and got 10997 )

Xbench (first number is stock 4GB, second is 16GB @ 1333mhz):
Overall Score: 334.66 / 329.49
CPU: 266.70 / 271.39
Memory Test: 633.23 / 625.53
Quarts Graphics Test: 526.67 / 505.27
OpenGL: 251.87 / 244.61
User Interface: 267.17 / 261.48


Handbrake encoding*:
Stock 4GB @ 1600mhz - 6:42
16GB @ 1333mhz - 6:43

* I ripped a copy of 10 Things I Hate About You in VOB format and converted it to h.264 at Handbrake's "Normal" Setting

One of two conclusions can be made of the above results:
1. In none of the tests was there a statistical significant difference in the results (meaning that the results were all to close to say either was "faster" than the other by any major margin).
2. If upgrading from the Stock amount of RAM, then going with "slower" RAM will not significantly impact performance. This can be because of one of the following:

A. The Mini is starved for RAM at 4GB and any loss in performance by using slower RAM is made up for by having more.
B. Or in the end the bandwidth loss by going with slower 1333mhz (but lower latency) isn't enough of a performance loss to make a significant difference in the over all performance of our Mini's.

So my point? Buy whatever is a good deal because in the end it probably isn't going to make a significant difference unless money is no object (of course).
 

paulrbeers

macrumors 68040
Original poster
Dec 17, 2009
3,963
123
What's the savings? Both are about $85.

What are your benchmarks for 16GB 1600?

Actually I picked up the 1333mhz ram for only $48! So for that price it was a no brainer, especially since the benchmarks show no loss in speed compared to stock 1600mhz @ 4gb
 

philipma1957

macrumors 603
Apr 13, 2010
6,365
251
Howell, New Jersey

Attachments

  • Screen Shot 2013-01-25 at 11.03.50 PM.png
    Screen Shot 2013-01-25 at 11.03.50 PM.png
    193.3 KB · Views: 268
Last edited:

philipma1957

macrumors 603
Apr 13, 2010
6,365
251
Howell, New Jersey
the fastest ram is this;

correction below
http://www.newegg.com/Product/Product.aspx?Item=N82E16820104317
price 125


paul used this

Both links in that thread are for 1600 RAM, and both are Corsair Value Select... That RAM is not Apple qualified/certified and is notorious for causing unstable behaviour with Macs.

the 1333 ram I have in the post above you is patriot it is good stable ram for 2011 mac minis and 2012 mac minis. it was on sale for 48 bucks.
 
Last edited:

53x12

macrumors 68000
Feb 16, 2009
1,544
4

philipma1957

macrumors 603
Apr 13, 2010
6,365
251
Howell, New Jersey
That's way overpriced at $130. There's no advantages to CAS9 in these computers. RAM running at 1333 isn't perceptibly faster; this would make less difference than that.

I never told any one to use it . I answered the simple question what is the fastest ram available for the 2012 mac mini . I don't use it as only a handful of people would be able to get any real world benefit from it.

A) you need to use more then 13gb ram often
B) you need to be doing ram intensive tasks


Basically any ram that works with a good price is what I use.
 

fhqwhgads

macrumors member
Jun 13, 2003
45
0
I guess maybe I'm confused, but doesn't having 4 times the RAM @ 1333 and getting almost no benefit vs 1600 show that the bus speed makes a big difference? Wouldn't the helpful test be to compare equal amounts of RAM at the two bus speeds? I look at your results and make the exact opposite conclusion: that having the decrease in bus speed means you need 4 times as much RAM to get basically the same performance.

Help me understand what I'm missing here as I'm definitely not super savvy about this sort of thing.
 

paulrbeers

macrumors 68040
Original poster
Dec 17, 2009
3,963
123
I guess maybe I'm confused, but doesn't having 4 times the RAM @ 1333 and getting almost no benefit vs 1600 show that the bus speed makes a big difference? Wouldn't the helpful test be to compare equal amounts of RAM at the two bus speeds? I look at your results and make the exact opposite conclusion: that having the decrease in bus speed means you need 4 times as much RAM to get basically the same performance.

Help me understand what I'm missing here as I'm definitely not super savvy about this sort of thing.

More RAM rarely means better benchmarks unless your computer is starved for RAM. More RAM is needed so you can run more stuff at any given time. So basically unless 4gb is too little to handle the OS + benchmarks, 16 vs 4 matters none. More RAM does not equal faster, it just means you can have more applications open before your system slows...

----------

So what is the best cheapest ram 16gb for the 2012 base mini ?

Can you post a link

You can use any sodimm 1333mhz or better DDR3. It doesn't have to be "Mac RAM". However, if you do buy "pc" RAM be sure to check the reviews for how it works in Macs. There is no difference between Mac and PC ram other than occasionally some of the PC RAM is out of spec for a Mac. With that said as of the Core i series, the memory controller is on the CPU now so there is very little RAM that won't work in a MAC that works in a PC. You can usually save yourself $10-20 by going PC ram
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.