SI-PEX40058 2 Port SATA III 2 Port eSATA III PCIe 2.0 x2 HyperDuo Controller Card Green

Had to give it a 3 star on stability work great on older systems but I tried to install in new Z590 mobo and it caused the system to not do anything not even POST so had to reinstall in older system and find a new one and that was not an easy task.

Worked out of the box in Windows 10.

Horrible. I can't believe this thing is being sold. There are so many things wrong with it I don't even know where to begin but I'll try. First let me say where I'm coming from. I had already read the reviews and didn't expect much in the first place, but I've been looking for an AHCI (specifically AHCI) chipset that implemented the FBS (FIS-Based switching) capability for a long time and these guys advertised it. Unfortunately this chipset barely implements enough of the AHCI spec for our AHCI driver to probe the disks. It only reports three capabilities. It reports AHCI v1.2 and S64A, NCQ, and PMD, and that's it. It does not report SSNTF or SCLO and it definitely does not report FBS (FIS-Based Switching support). Without SSNTF and particularly without SCLO there's no way this chipset can be even remotely reliable. It also doesn't report low-power or other mod e capabilities and is missing numerous other standard capabilities. For example, the motherboard AHCI chipset reports: S64A, NCQ, SSNTF, SMPS, SALP, SAL, SCLO, SPM, PMD, SSC, and PSC. (Obviously I'm testing on a Linux or BSD system and not windows, so I have full control over the driver and know exactly what the chipset is doing at the PCIe host interface level). If this chipset supported proper port multiplier detection it would report SPM (Port Multiplier support on each port) and if it supported FIS-Based switching via the AHCI port multiplier spec it would report FBS. It reports neither. What this chipset is doing is basically aggregating the real ports behind the scenes and doing the port multiplier detection and scanning behind the scenes, and reporting them as separate AHCI ports. In other words, it is hiding the port multiplier behind some firmware and chip magic rather than implementing the port multiplier spec on the host-facing ports. Honestly I'm not sure how it is supposed to work when the chip only reports 8 ports (one of which contains the MARVELL controller... also off-spec), is aggregating PM ports behind the scenes, but has two external and two internal SATA ports (which theoretically should be able to support at least 20 drives if one connected a port multiplier to each port). This might be OK except for three things: (1) Without SCLO you can't have a reliable AHCI chip, period. (2) If it is implementing FIS-Based switching internally it sure isn't showing up when I do read tests on the ports. With two hard drives in a port multiplier I get ~70MB/sec reading from either but if I read from both simultaniously I get ~35MB/sec from each. In otherwords, it is clearly NOT doing FIS-Based switching internally when you have an actual port multiplier connected to the E-SATA port. And (3) It doesn't handle hot-plug properly. When I booted the machine the first time it detected two of the three drives in my PM. When I hot unplugged and replugged two of them it detected that two of the ports unplugged and plugged, but then failed to detect one of the two replugged drives. In addition, for about 30 seconds before detecting one of the two replugged drives the console was spewing DMA engine failures. When I rebooted it got stuck in the BIOS (didn't even get to the machine boot)... twice. A few reboots later it got past the BIOS and reported all three disks. It's a real mess. Not even remotely reliable. This chip is basically trying to do PM detection behind the scenes and presenting the individual drives as individual AHCI ports to the host, instead of implementing the port multiplier AHCI chipset feature and exposing it to the host. And it's doing a horrible job of it. I can see why others have reported Windows BSODs with this chip family. The firmware is insanely bad, even worse than the SiS cards. I was hoping that it implemented the port multiplier AHCI spec. It doesn't. I was hoping that it implemented FIS-Based switching... it really doesn't seem to. I recommend that people give anything based on this chip family a very WIDE berth. Avoid it at all costs. It's total junk. -Matt

Installation of the SI-PEX40058 card was easy. Windows Home Server 2011 automatically found and installed the drivers without any problems. So far throughput seems to be very good although I have not ran any technical speed tests. Of minor concern the IO Crest box looked like it had been opened before although all items were included.

Tried it out in my Mac Pro 4,1 - bootable, but causes Kernel Panic when waking from sleep. Also, drives connected to the internal SATA ports show up as external in OS X 10.8.4. Black Magic's disk speed tool reports 220MBps write and 440MBps read - a bit worse (especially the write) than the SATA III via Intel Z77 through Windows with the same drive (Sandisk Extreme 240GB). Returning and trying an Apricorn Solo X2.

1 port is dead. I use the other port. Adds 20 seconds to my boot time. Sleep doesn't work with Windows 7 booting from SSD attached to this device. Have to use the much slower hibernate. Device speed is as advertised.

I wanted to add an SSD Drive to a Dell 3000s Slim PC that I plan on using as a home theater PC. The motherboard comes with only 2 SATA ports and room for a DVD Drive and one 3.5" Hard Drive. I needed to add a port to install the SSD. I also wanted eSATA ports for possible future expansion. My plan was to use this card with the existing HD and attach the new SSD boot drive to the motherboard header. Worked like a charm. First I cloned the existing hard drive to the SSD. Then, I had to change the bracket on the card to the low profile bracket that was in the box. Next, I plugged in the card into the open 16x expansion slot (it would not work with the available 1x slot). Then I added an SATA power splitter so I could power the new drive. Next, I added the SSD to a small drive rack I velcro'ed to the bottom of the PC and replaced the existing 1 TB HD with a new 2 TB WD Red HD I had laying around. Finally I reconnected the drives to the SATA ports and booted the PC. The add-in card was immediately recognized and the drivers installed by Windows 8.1. I had to go into the Disk Management Snap-In Console to initialize and format the 2 GB drive but after that it seems to be working fine. Now the PC boots in about 7 seconds vs the original 45 seconds, everything is so much snappier and I have more than double the storage space. I still need to try the external ports. Will do that soon. To summarize, I'm not using this card in an advanced way or using anything beyond its basic features and I don't plan to. I'm not concerned with the fastest speeds or using it with a RAID setup. It just needs to be fast enough for my relatively slow 5400 RPM hard drive(s) to record and playback my TV shows and be ready for the possible addition of future external drives. I deducted one star because the board is a little expensive for what I needed it for ($50) but it was the only one that I could find locally with the ports I wanted that fit in a low profile case. If I could have deducted only half a star I would have.

Plugging in the PCIe card and attempting to do a Windows 7 install causes a BSOD (even with no hard drives connected to any of the SATA ports). Removing the PCIe card and attempting to do a Windows 7 install DOES NOT cause a BSOD. Therefore, this card is causing a BSOD.