The SSD is a SATA drive just like the HDD, and shouldn’t require a special driver.
Agreed. My laptop multi-boots 7/10/XP on a SSD without any problems. I'm using truly independent OS partitions, not a Microsoft-style BCD-tangled multi-boot ... though I don't think that matters (other than a truly independent OS being easier to troubleshoot).
Yet, I can sympathize with johnhoh's skepticism because you've tried directly cloning partitions without resizing, so there's not much left in the way of uncontrolled variables. I can think of only three that haven't been definitively ruled out yet by your tests:
- HDD vs SSD (johnhoh's argument) ... it shouldn't matter, but we're past the point of assuming anything by now. I don't know if you have a spare HDD laying around, but if so it might be helpful to try cloning from HDD to HDD -- if nothing else, just to rule out this particular SSD as being an issue. (That is, assuming HDD-to-HDD also fails, which I expect it would.)
- Choice of cloning program ... as I've mentioned earlier, IME image/restore tends to be slightly more robust than direct cloning, but you said you can't do that. I believe you've tried the direct cloning with Macrium Reflect, though, which IME has been one of the more reliable programs I've come across, so I'd be surprised if something else could do this if Macrium can't.
- This is kind of a long shot, but some motherboards initialize HDDs (and SSDs) with different geometry (240 tracks/cylinder instead of the more common 255 tracks/cylinder). I'm not sure about 7, but I know this could cause problems with XP. It might be helpful to use a diagnostic tool to determine the geometry the SSD has been initialized at.
To expand on that last point, those "tracks" or "heads" per cylinder numbers are all fictitious (and have been since the advent of IDE controllers some 25 years ago), but even though they're made-up numbers they've been known to trip up the retro bootloader in earlier OSes like 98, 2000, and XP. Some laptops -- notably, Compaq, IBM and Lenovo -- used 240 heads as the default while Dell and most others used 255 heads. Using the wrong geometry would cause the boot process to fail.
It's worth noting that a difference here could affect direct cloning because your target disk is external. The internal disk's geometry is set by the motherboard, but the external disk's geometry is determined by the USB-to-SATA translator. An internal source disk could be seen as having 240 heads while an external USB disk might be seen as having 255 heads. Cloning in such a scenario could result in boot failures.
In contrast, this issue does not affect the image/restore technique because by the time the image is restored the target disk will be installed *internally* in the same place the source disk was, so will be seen with the same geometry. An image contains no geometry info and the geometry of the restoration is determined at the time the image is restored.
(This is one reason why I argue that image/restore is more reliable than direct cloning -- the USB-to-SATA translation blinds the cloning tool to the geometry of the target disk.)
I've been assuming MSI is one of the "others" and thus defaults to 255 heads, so my reasoning might be a shot in the dark ... but again, we're past the point of assuming anything so it might be worth checking out.
I'm (on vacation at the moment and) not at my computer where I can check right now, but perhaps someone else can recommend a suitable diagnostic tool to determine what geometry is in use by the SSD. Off the top of my head, I think Cliff Greiner's Test Disk is one such tool. Install the source disk and check the geometry, then install the target disk and compare results.
, plug in the original HDD via the SATA-USB adapter (the old XP partition would then appear as drive G
