Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
@shortcut3d

Did you happen to take scores with 1333 ram before testing the 1866?

Or does anyone know of posted scores from others who have used 8GB, 1333 Ram on a similar system? Bootcamp native only.

I'm curious as to how the scores may differ. Or possibly differ between 1600 vs 1866 too.

@philipma1957

The way I installed Windows 7 was simply by using a USB drive, having bootcamp convert the ISO, and then installed the bootcamp drivers at the end. I also did the install with ethernet hooked up to begin with, just to simplify internet connections.

I like VMware player on the PC side of things. Also helps that it's free (weird how they charge for the Mac version).

Virtual Box is also okay for anyone out there who wants something free, for both Mac and PC, and doesn't have VMware or Parallels. But back to testing... when I tested Win 7 under Parallels, my graphics scores were miserable. Not that they are great anyway, but they were considerably worse.

Ram-wise, I've started to see some negative reviews on the Corsair Vengeance as far as actually bypassing the bios. I don't believe anyone has tried it on a mac mini yet (according to reviews), but I'm not feeling super positive about it. Sort of waiting for a sale in HyperX, or taking the plunge on another brand. Ideally I'd like to find some before and after Win 7 scores (native install, same mini), to see which speed (and brand) may be best. But I don't think anyone out there has done in-depth testing.
 
Guess you can be the guinea pig and tell me if it actually works? *grin*

On some of the 4GB (2x for 8GB) chips, there are reviews stating it's not showing up at 1600Mhz. But I'm not sure what machines they were using.
 
@shortcut3d

Did you happen to take scores with 1333 ram before testing the 1866?

Or does anyone know of posted scores from others who have used 8GB, 1333 Ram on a similar system? Bootcamp native only.

I'm curious as to how the scores may differ. Or possibly differ between 1600 vs 1866 too.

@philipma1957

The way I installed Windows 7 was simply by using a USB drive, having bootcamp convert the ISO, and then installed the bootcamp drivers at the end. I also did the install with ethernet hooked up to begin with, just to simplify internet connections.

I like VMware player on the PC side of things. Also helps that it's free (weird how they charge for the Mac version).

Virtual Box is also okay for anyone out there who wants something free, for both Mac and PC, and doesn't have VMware or Parallels. But back to testing... when I tested Win 7 under Parallels, my graphics scores were miserable. Not that they are great anyway, but they were considerably worse.

Ram-wise, I've started to see some negative reviews on the Corsair Vengeance as far as actually bypassing the bios. I don't believe anyone has tried it on a mac mini yet (according to reviews), but I'm not feeling super positive about it. Sort of waiting for a sale in HyperX, or taking the plunge on another brand. Ideally I'd like to find some before and after Win 7 scores (native install, same mini), to see which speed (and brand) may be best. But I don't think anyone out there has done in-depth testing.

Not quite a before and after. The ram quantity should not affect the test results because Geekbench results only measures the "necessary" amount of memory. After 8GB you will not see any improvement in scores.

This test is the 16GB 1333MHz Corsair Value Brand in the same system:

http://browse.geekbench.ca/geekbench2/528108

FWIW, I get almost the same variance with a clean install of Windows 7 x64 vs an install after typical use, as I get going from 1333MHz to 1866MHz.
 
I just ordered Corsair Vengeance 16GB (2 x 8GB) 1600MHz memory. This will likely end up in my 2011 15" MacBook Pro since my use case for the 2011 Mac minis do not utilize that much memory.

I, too, am planning to upgrade my 2011 Mac Mini's RAM from 4GB to 8GB, and this discussion about supporting a RAM of higher speed really piqued my curiosity. I really doubt that a 16GB kit will give me good RoI, but a 8GB kit at higher clock speed just might. It would be great if you could test out the RAMs on the Mini and tell us if their native clock speed is detected by the Mini.
 
I, too, am planning to upgrade my 2011 Mac Mini's RAM from 4GB to 8GB, and this discussion about supporting a RAM of higher speed really piqued my curiosity. I really doubt that a 16GB kit will give me good RoI, but a 8GB kit at higher clock speed just might. It would be great if you could test out the RAMs on the Mini and tell us if their native clock speed is detected by the Mini.

The clock speed will adjust if the memory is PnP (Kingston) or Auto-detect (Corsair). Basically, you have to read the detail that says it will adjust without entering the BIOS on Sandybridge Intel i3, i5, i7.
 
The clock speed will adjust if the memory is PnP (Kingston) or Auto-detect (Corsair). Basically, you have to read the detail that says it will adjust without entering the BIOS on Sandybridge Intel i3, i5, i7.

I did run across a couple of reviews for the corsair saying that it didn't run at 1600 for them. But... I don't think they were using macs, and perhaps weren't even using sandybridges.

If you don't mind, can you test in a mac mini (i5) when you get the ram, and report back if it registers at 1600? That's the same ram I am thinking of getting, but in 4GB (2x8GB) form. Decent timings and price for that brand, at least compared to the HyperX.
 
Wirelessly posted (Mozilla/5.0 (iPhone; CPU iPhone OS 5_1 like Mac OS X) AppleWebKit/534.46 (KHTML, like Gecko) Version/5.1 Mobile/9B179 Safari/7534.48.3)

namelessme said:
The clock speed will adjust if the memory is PnP (Kingston) or Auto-detect (Corsair). Basically, you have to read the detail that says it will adjust without entering the BIOS on Sandybridge Intel i3, i5, i7.

I did run across a couple of reviews for the corsair saying that it didn't run at 1600 for them. But... I don't think they were using macs, and perhaps weren't even using sandybridges.

If you don't mind, can you test in a mac mini (i5) when you get the ram, and report back if it registers at 1600? That's the same ram I am thinking of getting, but in 4GB (2x8GB) form. Decent timings and price for that brand, at least compared to the HyperX.

Unfortunately, I only have the systems in my signature which are all Intel i7 (1 Clarksdale and 3 Sandybridge).

My understanding is the Kingston Hyper X timing is better at 1600MHz than the Corsair Vengeance. Although I may be wrong.
 
I'll be the guinea pig for the 8GB Corsair vengeance in a base model mac mini. I just ordered it from Amazon, so hopefully sometime next week I get it in and I'll run some tests.

Price is decent at Amazon, at around $44 for 8GB, which isn't that much more than 1333 Mhz speed Ram. Amazon also has a good return policy, in case it doesn't work.

It also helped that Amazon has a promotion today at Amazon Local, $5 for a $10 gift code, and I also previously converted a $5 best buy reward into Amazon credit... and had $3 credit for some old promotion... and also found out that Amazon can use Discover rewards (had $7 in there for like the last 5+ years) -- ended up costing about $27 total.
 
Wirelessly posted (Mozilla/5.0 (iPhone; CPU iPhone OS 5_1 like Mac OS X) AppleWebKit/534.46 (KHTML, like Gecko) Version/5.1 Mobile/9B179 Safari/7534.48.3)

I tested the Corsair Vengeance 16GB 1600MHz kit in my 2011 MacBook Pro overnight and it appeared to be stable. No errors in Rember and passed the Geekbench stress test for 20+ loops. I really like the kit, but even with better CAS timing the gains are not as high as 1866MHz kits. I'm seeing double the gains when comparing the two.
 
Did you see some improvements when comparing the Vengeance to 1333Mhz Ram, at least?

I would expect the 1866 to be better, but I too had hoped the lower timing would give a bit of a boost to the 1600 Vengeance.

So long as I get some improvement over 1333Mhz Ram, I'd be happy.
 
The Sandy Bridge chips get virtually ZERO benefit from RAM faster than 1600mhz. There is no point in wasting your money. It might give you a tiny advantage in 3D graphics if you don't have the Mini with dedicated graphics, but that's it. Often the difference is 1% or less. Just grab a good 8GB kit of 1600 and call it a day, or heck, even the difference between 1600 and 1333 is slight.
 
Just updating things in case anyone else was interested in trying Corsair Vengeance in a base model Mac Mini.

I got the Ram today, installed it, and it appears to be running at 1600 Mhz. At least it's listed at 1600 under the preferences under Lion.

I just ran some quick tests now in Windows 7. If anyone knows if these are decent or not, let me know, as I have nothing to really compare to (besides much worse memory and graphics scores when using 2 GB).

Win 7 index:
System Score: 5.9
Memory Score: 7.5
CPU Score: 6.9
Graphics Score: 6.3
Gaming Score: 6.3
Disk Score: 5.9

I also ran Geekbench, but could do the 32 bit test only, due to not owning the full version:

Geekbench 2.2.7 for Windows x86 (32-bit)
Integer - Processor integer performance 4962
Floating Point - Processor floating point performance 6052
Memory - Memory performance 6716
Stream - Memory bandwidth performance 8091
Geekbench score: 6007

I ran it twice... noticed slightly higher numbers the second time. And under Geekbench for Ram I also get this: 8.00 GB DDR3 SDRAM 799MHz
Which I assume we just 2x for the 1600 speed.

So... do the numbers look right to people? I've never been a big benchmark person, so don't know if the numbers are correct, or decent... or if ram looks speedy.
 
Your Windows Experience scores look accurate. My 2.7GHz Dual-Core Intel i7 has 7.6 memory with 1866MHz, whereas my 2.0GHz Quad-core Intel i7 has a perfect 7.9 with 1866MHz.

There another thread about building the fastest 2.7GHz Dual-Core Intel i7 Mac mini. You should reference it for additional benchmarks.
 
Thanks for the reply. Started looking at other benchmarks...

If accurate, I think my geekbench bandwidth numbers look good. I also checked the Ram you used, 16 GB of vengeance at 10-10-10-27, right?

I used 8GB 9-9-9-24. Could the lower timings explain why my bandwidth numbers look good? Or am I getting confused as to what the numbers even mean? I assume higher is better for geekbench scores.
 
Wirelessly posted

CAS Latency will have a small impact in performance. Lower CAS and RAS timings are better.
 
Just updating things in case anyone else was interested in trying Corsair Vengeance in a base model Mac Mini.
Hey man, could you be so kind and share the link to the Amazon item you bought?
I'll basically order the mini in the next days and will obviously order RAM as well, so since yours seems to be just fine and dandy...
 
The current Sandy Bridge architecture on the Minis doesn't really take advantage of faster RAM. Even in the most extreme case you get maybe 2-3fps more during a x264 first-pass encode, and under 1fps increase in game performance.

Source (tested with a i7-2600K): http://www.anandtech.com/show/4503/sandy-bridge-memory-scaling-choosing-the-best-ddr3/8

Sandy Bridge's improved memory controller has all but eliminated the need for extreme memory bandwidth, at least for this architecture. It's only when you get down to DDR3-1333 that you see a minor performance penalty. The sweet spot appears to be at DDR3-1600, where you will see a minor performance increase over DDR3-1333 with only a slight increase in cost. The performance increase gained by going up to DDR3-1866 or DDR3-2133 isn't nearly as pronounced.
 
Hey man, could you be so kind and share the link to the Amazon item you bought?
I'll basically order the mini in the next days and will obviously order RAM as well, so since yours seems to be just fine and dandy...

It was this one: http://www.amazon.com/gp/product/B006EWUOL8/ref=oh_o00_s00_i00_details

Corsair Vengeance 8 GB (2x4GB) DDR3 1600MHz PC3204 Pin SODIMM Laptop Memory CMSX8GX3M2A1600C9

currently at $43.99, which is only a little bit more than 1333 Mhz Ram usually goes for. And one neat thing Amazon lets you do is also use rewards pts from some credit cards during your purchase, like I mentioned above for Discover -- even if you don't have enough to cash them out normally -- sort of like a free discount then.

The Ram itself seems to be chugging along fine so far. But I haven't had time to do extensive ram tests on it yet. I did do three passes of Rember with no problems though. As for those who keep commenting that Sandy Bridge doesn't take advantage of speedier Ram... your own quotes from the posted article state that 1600 Mhz is the sweet spot, which is what this Ram is.
 
As an Amazon Associate, MacRumors earns a commission from qualifying purchases made through links in this post.
Wirelessly posted

Tastydirt said:
The current Sandy Bridge architecture on the Minis doesn't really take advantage of faster RAM. Even in the most extreme case you get maybe 2-3fps more during a x264 first-pass encode, and under 1fps increase in game performance.

Source (tested with a i7-2600K): http://www.anandtech.com/show/4503/sandy-bridge-memory-scaling-choosing-the-best-ddr3/8

Sandy Bridge's improved memory controller has all but eliminated the need for extreme memory bandwidth, at least for this architecture. It's only when you get down to DDR3-1333 that you see a minor performance penalty. The sweet spot appears to be at DDR3-1600, where you will see a minor performance increase over DDR3-1333 with only a slight increase in cost. The performance increase gained by going up to DDR3-1866 or DDR3-2133 isn't nearly as pronounced.

FWIW, 2-3fps in Handbrake is a huge gain from memory since the x264 encode is not that memory intensive, but processor dependent. At this point the price difference between 1333, 1600, and 1866 is very small.
 
Has anyone tried the Corsair Vengance 8GB 1866MHZ (CMSX8GX3M2A1866C10)? (for about $90.00 USD)
Amazon Link

specifically I'm thinking of putting it in the Mac Mini Server (quad i7 2Ghz) somewhere I saw that the faster speed would provide some benefit to the GPU, and since Apple doesn't offer descrete graphics for this machine I'm wondering if this would provide some benefit?

Cheers,
~Ark
 
As an Amazon Associate, MacRumors earns a commission from qualifying purchases made through links in this post.
Has anyone tried the Corsair Vengance 8GB 1866MHZ (CMSX8GX3M2A1866C10)? (for about $90.00 USD)
Amazon Link

specifically I'm thinking of putting it in the Mac Mini Server (quad i7 2Ghz) somewhere I saw that the faster speed would provide some benefit to the GPU, and since Apple doesn't offer descrete graphics for this machine I'm wondering if this would provide some benefit?

Cheers,
~Ark

It's down to$ 79.99 Now, that was fast.


Also, a general question.

How much performance difference would there be in a Mac Mini Server (mid 2011, Quad i7) in these options:

* Speed: 1866MHz, Timing: 10-10-10-27, 1.50v (08 GB) Corsair/Kingston
* Speed: 1600MHz, Timing: 09-09-09-24, 1.50v (08 GB) Corsair/Kingston
* Speed: 1600MHz, Timing: 10-10-10-27, 1.50v (16 GB) Corsair/Kingston
* Speed: 1333MHz, Timing: CL = 9, 1.35v (16 GB) Crucial
* Speed: 1600MHz, Timing: CL = 11, 1.35v (16 GB) Crucial
* Speed: 1333MHz, Timing: CL = 9, 1.50v (08 GB) Crucial
 
Last edited:
As an Amazon Associate, MacRumors earns a commission from qualifying purchases made through links in this post.
It's down to$ 79.99 Now, that was fast.


Also, a general question.

How much performance difference would there be in a Mac Mini Server (mid 2011, Quad i7) in these options:

* Speed: 1866MHz, Timing: 10-10-10-27, 1.50v (08 GB) Corsair/Kingston
* Speed: 1600MHz, Timing: 09-09-09-24, 1.50v (08 GB) Corsair/Kingston
* Speed: 1600MHz, Timing: 10-10-10-27, 1.50v (16 GB) Corsair/Kingston
* Speed: 1333MHz, Timing: CL = 9, 1.35v (16 GB) Crucial
* Speed: 1600MHz, Timing: CL = 11, 1.35v (16 GB) Crucial
* Speed: 1333MHz, Timing: CL = 9, 1.50v (08 GB) Crucial

this all depends on how you want to use your server. 16gb comes in handy for users of vmware fusion people that have a mini with 1 lion osx and 2 windows virtual machines need 16gb ram.

if you don't need 16gb ram and don't get pageouts with 8gb ram.

the kingston 1866MHz is fastest. also if you need 16gb ram because you do virtual machines here is a good sale at newegg it ends in a few hours

http://www.newegg.com/Special/ShellShocker.aspx?Tpk=shell shock


85 dollars for 16gb of ram that will work with the server mini sale ends in 2 hours 4 east coast time.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.