As mentioned in previous threads, shortly I'm going to be transplanting my home server into a new case and I promised to take some pics as the case is still fairly new and there's not much info about it just yet. So, enter the InWin MS08 server tower case which'll be the primary focus of this post. InWin IW-MS08 It's a fairly compact case and as can be seen from the spec sheet it takes mATX and mITX motherboards, 1U or FlexATX PSUs, has eight hot-swappable 3.5" SAS/SATA bays and provision for up to four 2.5" discs - more on that in a bit. Cooing comes in the form of two 80x25mm PWM fans (from a company called NMB) attached to the rear of the hot-swap backplanes; exhaust is two more of the same 80x25mm PWM fans at the rear of the case. PCIe slots are full-height with a quick-release clampy thing if you don't know how to use a screwdriver. General impressions so far are good considering the relative cheapness. The steel used for the chassis isn't the thickest, but there's no undue flex and it all fits together very nicely - no percussion required to get the side panels back on. The locking plastic facia on the front wouldn't stop anyone remotely determined, nor would the small lock on the side do much to prevent anyone from opening the case (which would in turn allow you to unclip the plastic facia anyway). Anyway, enough of the nonsense, let's pop open the case and get on with the pics. I'm only bothering to post ones of the interesting bits inside, but let me know if there's anything else people want to see, or further information on. Exterior of the optical bay, 2.5" SATA hot-swap caddies and the top set of 3.5" caddies. At first I thought the 2.5" cage area was just going to be empty, since the available pics showed a blanking plate and the listing showed the cage as "optional" - so I figured that the cage didn't come as standard and I'd end up buying it later when it eventually made its way to the UK as a massively overpriced optional extra. Imagine my surprise when I open the box and see it already installed. So I've no idea on the part number and I've no idea if the cage is actually standard or whether or not it will come with the case you might end up buying, but it came with mine and this is what it looks like. Note that the 3.5" drive trays are the same as used on my other InWin chassis, and have the same hexagonal locking device - although oddly enough the hex key wasn't supplied with this case. View of the backplane with the lower four 3.5" drive trays removed: View of the backplane with the two 2.5" drive trays removed. Note there's precious little ventilation here (nor any active cooling), and the serial sticker on the reverse is blocking half of the ventilation holes so I wouldn't recommend using this for drives that may get very toasty like 10k spinners. Time to pop open the case... Detail of the plastic quick-release PCI retention bracket. There's a small fold-out loop for use with a lock just above it. Detail of the rear of the 2.5" hot-swap caddy; molex, two SATA plugs and some named jumpers. The first 7-pin one is labelled SGPIO so it seems reasonable to assume that if you've got a SAS -> SATA cable with SGPIO sideband, it'll work when plugged into an HBA. I've no idea yet whether there's any way to get an SGPIO output from onboard SATA on Supermicro motherboards - I suspect not but if anyone knows different please let me know. The second is a 2-pin named "signal" - at a guess that's going to be for drive activity LEDs. As mentioned previously, that large sticker is blocking ventilation holes. At the bottom right of the PCB obscured by the installation bracket is a 3-pin jumper called "Enable SGPIO" (with the sticker also obscuring this label). The internal 2.5" trays. The cage can be unlocked with a captive thumbscrew whilst the drive mounting uses only two screws on the bottom side, with lugs for the top. Vibration might therefore be a problem with platter-based drives but it'll be perfect for SSDs. Side-mounting screw holes are also a must in order to fit the lugs (but there are screw holes on the bottom and sides). Close-up of the rear of the backplanes at the back of the two sets of four bays. As you can see the PWM comes straight from the backplane itself, there's also a beeper (presumably for fan failure?) and an enticing-looking four pins which I'm surmising might be to plug the motherboard in to the PWM circuitry. But there's no proper manual or techie stuff on InWin's website any more (grrr) nor any special fan cables supplied with the case so it looks like these fans are on their own - and the first set likely to be replaced with Noctua's (likely NF-A8 PWM) if these NMB fans turn out to be loud. I'm more wary of using Noctua's for the rear fans because they don't seem to get on well with Supmicro's IPMI, but that's something to test further down the road. On further examination there was some enticing text written on the backplane PCB so I decided to see how easy it was to dismantle - very, as the case turned out. Undo the two top screws and the fan mount unhooks from the bottom, then two more screws and the PCB comes off. First the boring bit - four (SAS-compatible by the looks of it) HDD ports and part number 3RAMVI006100. Now for some juicy close-ups of the business end of the PCB. A manual for this would have been nice but there doesn't seem to be one available, so I've transcribed the text description to make stuff more search-engine friendly. Top half first, we have five jumpers running along the top. From left to right, here's the jumper names, function and help text. Values with an asterisk denote the current OotB setting and I assume that "En" means "Enable" and "Dis" means "Disable". Code: JM2 - SGPIO, pins 1-3 JM2 Function *1-2 Short SGPIO_0 En 2-3 Short SGPIO_0 Dis JM3 - UART1, pins 1-3 No help text, no pins jumped CN2, pins 1-6 CN2 Functions 1-2 Short Max fan rpm 60% 3-4 Short Max fan rpm 80% *1-2 and 3-4 Short Max fan rpm 100% *5-6 Open Temperature alarm 45C 5-6 Short Temperature alam 55C JM1, pins 1-6 JM1 Functions, no pins jumped 1-3 Short Master 3-5 Short Slaver 2-4 Short Test En 4-6 Short Test Dis JS1, pins 1-4 JS1 Functions, no pins jumped 1 Fail LED1 2 Fail LED2 3 Fail LED3 4 Fail LED4 Good to see there's decent enough fan control options at least. I will see how noisy these stock 80mm's are first before thinking about replacing them with Noctua's. On to the bottom half, here we can see a Nuvoton M052LBN ARM microcontroller responsible for looking after the backplane. There are three data lines into this chip coming straight from the SFF 8643 connector so it's safe to assume these are probably the in-band SGPIO lines. Beneath that, we've got a piezo speaker (which I assume will go biddly beep-beep whenever a fan or temp warning goes off), a PWM fan header and two more jumpers. Code: J1 - ICE, pins 1-5 No help text, no pins jumped CN1, 7 pins CN1 Functions 1 Fan fail LED + 2 Fan fail LED - 3 Temp Fail LED + 4 Temp fail LED - 5 Mute Sw + 6 Mutw Sw - 7 NC 8 Key pin (unused) As and when I get the server rebuilt/transplanted I'll see if the SGPIO works - soon to be plugged in here is a trusty M1015 reflashed to a 9211-8i, and I'm not sure if it supports SGPIO. I do have an IBM M1215 lying around somewhere - this hasn't been reflashed yet but IIRC it does support SGPIO, but sadly I don't have any spare 8643->8643 cables. As an aside, do all 8643->8643 cables include SGPIO? If so I'd be tempted to get a couple and see how the SGPIO behaves with the M1215 HBA. All three types of the 2.5" and 3.5" trays removed from the case. Note that the 3.5" trays do not have any mounting screws on the sides - screws are from the bottom only. The 2.5" trays should be tall enough to accommodate 12.5mm and 14mm height drives (although I've only got 9.5mm SSDs to test with). The same but with drives put in. Note the venerable C300 SSD used in the 2.5" hot-swap tray is a 9.5mm drive and there seems to be enough headroom for ~15mm drives although I don't have one on-hand to test.