There are many Mac computers with Nvidia MCP79 chipset (Nvidia MCP79 is present in many models of iMac, Macbook, Macbook Pro, Macbook Air).
Nvidia MCP79 have SATA II controller.
There is well known bug with SSD based on SandForce SF-2281 and any Mac with Nvidia MCP79. SSD negotiated Link Speed is 1.5 Gigabit (SATA I). But it must be 3 Gigabit (SATA II).
OCZ already have fix for this issue (for their SSD):
Corsair works on this issue (they have SSD based on SF-2281).
Patriot Memory works on this issue (they have SSD based on SF-2281).
Intel 520 SSD have same issue.
Intel will you fix this? Talk to Apple or release fix by yourself, but you must do something, because Intel 520 SSD with 150 MB/s read/write speed on Mac (with Nvidia MCP79) is indecent.
Thank you all for posting,
I am sorry for the inconvenience.
Please note that the compatibility between different devices including (but not limited to) SSDs and Apple* products is very specific. Therefore, we recommend you to contact Apple* for a list of validated Intel® SSD models that can be used with their products.
Will the Intel® SSD work in my MacBook?
This problem seems to be more widespread: Intel 330 in a Mac Pro with Intel ESB2 Chipset only negotiates on 1,5 Gigabit as well. I am quite disappointed, will probably return the drive
I too am experience this issue with an Early 2009 MBP 17" and the MCP79 chipset using the Intel SSD 520 480GB. It should be able to negotiate to 3.0Gbps with that SATA II interface, but as the OP has noted, this has plagued other drives with the Sandforce chipset along with this MCP79 interface.
It would be nice if Intel could release a "fix" like one of the other OEMs has done. Their solution was a simple Linux utility that forced the drive to a particular speed.
Come on Intel, please issue us a fix/workaround for this issue!
Im having the exact same problem! Currently running an Intel 330 inside of an Macbook Pro 5.1 (Late 2008) with Nvidia MCP79, and the negotiated link speed is 1,5 GBit while it should be at 3 GBit. My system is also plagued with beach balls every 3 minutes, and then the system is frozen for exactly 30 seconds.
Please Intel, fix this! At least tell us if there is a fix in the works, and an ETA when it will be available. Thanks!
(P.S. Has anyone tried running the OCZ tools to force the 3 GBit link speed in an Intel drive with sf-2281? Just curious.)
Update: it seems that my problem with the intel 330 and a 2008 mac pro follows a certain pattern:
- when cold booting the machine, i get the 3gbit connection
- when rebooting, i only get 1,5gbit
i then ordered a sata 3 controller for my machine, based on a asmedia 1061 chipset, and thought that the issue would be resolved when using native 6gbit. when connected to this controller, the intel 330 ssd always connects at 1,5 gbit, no matter if i cold boot, or if i reboot.
would it be possible to get some intel tool that would be able to lock the sata speed of the ssd to a certain setting?
I have the exact same problem with my Intel SSD 330 and MacBook 6,1 (2009 unibody).
Seeing that other manufacturers are actively trying to resolve this issue, I hope Intel will soon follow suit.
I've got the same problem and it's driving me nuts. My MacBook 5.1 late 2008 is beachballing every few minutes or so for around half a minute. I trusted intel because of their name and I probably wouldn't have gone with their SSD if they had not been giving so much praise for their reliability. I'm very disappointed and will probably never buy one of them again. I was sure that they would fix this in no time, but it's been 3 months now! What the heck.
I have the same problem. It seems like this is not a priority for Intel, as such a substantial time has now passed since the launch of the product. I will never buy an SSD from Intel again as the general device compatibility support is inferior to, for example, OCZ.
Intel have answered me. Intel will not fix that issue. So any users with Mac (Nvidia chipset) SHOULD not to buy Intel SSD based on Sandforce.
I bought this product immediately after release, tested specifically for nVidia SATA2 compatibility (and confirmed it was broken, locking down to SATA1) and reported the bug to Intel. They were nice enough about responding to it, but they have done nothing to resolve it. This is nothing new for Intel, the 510 had the same problem and they never resolved it either. The 510 also won't perform to spec on AMD AHCI drivers, only msahci drivers which I have confirmed on multiple SATA3 platforms. My general rule with newer gen SATA3 Intel SSDs is that I will only use them on Intel SATA chipsets, because they have proven to be marginal in both performance and compatibility on non-Intel platforms. This type of firmware support is certainly not worth their price premium.
I run a 320 in the same system and it works perfectly with both SSC enabled and disabled. Some SSDs will work properly on nVidia SATA2 at full speed with SSC disabled, but not with it enabled--but a large portion lock down to SATA1. Most systems have no access to enabling or disabling this feature (Spread Spectrum Clocking), so this is basically a responsibility of manufacturers at a firmware level.
This is not a rare problem; a Samsung 830 I tested last night was unstable in either mode, and is basically unusable. A Kingston V200 wouldn't even detect until I used a different cable, but it also locked down to SATA1. The 320 worked perfectly with all cables, in all modes.
As much as these newer Intel SSD models are faster, in my mind nothing Intel has produced has been as refined of a product as the 320. Since I haven't seen them do anything to improve or fix these problems, I don't really buy any more 520s since finding this bug & lack of legitimate action to fix, and I probably won't buy 330s either as it's just another twist on the same architecture. I prefer using SSDs that get firmware updates like the rest of the industry (or that have utilities to work around these type of problems), as LSI/Sandforce themselves will probably fix this eventually. I'm an IT consultant who implements a lot of these.
Recent Plextor products with current firmware sync properly on these platforms with SSC disabled; I reported the bug to them and they fixed it which I was impressed by. The M4 with their latest 000F firmware locks down as well in either mode on my test platform (non-Mac, ASUS AM2 motherboard).
hello. What is the status for this situation? Has anyone from intel responded? We need an update. Intel should provide a fix for this so we can enjoy the product at sata 2 speed. Currently I have sata 1.
I bought two Intel 330 SSD drive one for my iMac and the other for my Macbook pro. All of them works in 1.5MBps speed. Have to have a fix for it. It is very disappointing.
Same problem here, just fitted an Intel 520 in a MacBook Pro mid 2009, only 1.5 Gbps speed. Very strange Intel has no fix for this. Can i give the drive back?
This Nvidia SATA chipset (or any other SATA chipset made by Nvidia) is obsolete, no longer in production, or sold by Nvidia. Why Apple used them at all is astonishing to me, as they were on there way out before 2009. Their reaction to receiving data that they are connected to a SATA III device, which they don't understand, is to shift into "safe mode", or SATA I, since they apparently don't even try to negotiate to SATA II. Their support for the standard SATA protocols was problematic when they were in production, and Nvidia left that segment of the chip market behind them.
A link to OCZ's apparent fix for this problem is included in the OP, but that is only for a subset of their products, not all of them. If you check that link, you learn the fix must be applied by the owner, it is not done at the factory as a standard update, which implies it is an either-or fix, change it for Nvidia chipsets, or not. I also found that they downplay its availability, since when used incorrectly, it causes new issues for some users (a way of not saying what I just wrote.)
Why can't Apple or Nvidia supply an update to their products to support current technology? We already know they won't, so why is Intel wrong for not providing special support for a product they have nothing to do with, is no longer produced, and is well known to be a poor performing, buggy chipset?