Skylake-Based Z170 Gaming Mini ITX Motherboard

Really?
So this article and pic is wrong then?
block%20dia.png
Please be clearer when you try to refute someone's argument, I have no idea what you are referring to what I might be wrong about.

Why I commented on you earlier, is because of the following comments:
The asrock shuts down sata ports instead of the pcie slot. The Asus leaves the sata ports alone but takes away pcie lanes, gotta pick your trade off depending on board.

From the Asrock documentation
Please be noted that if the Ultra M.2 Socket (M2_1) is occupied by a SATA-type M.2 device, SATA3_4, SATA3_5 and the SATA function of the SATA Express connector will be disabled

You then replied:
Depends if you have AHCI or NVME, Divirge probably has AHCI and that will not be correct on a NVME.
His motherboard might not (most probably won't) be x16 with NVME, which is what most people will get.
This is not true, AHCI or NVMe have nothing to do (directly) with which ports or lanes are disabled.

AHCI or NVMe don't decide or influence if the PCIe x16 slot gets to keep all 16 lanes since it's a protocol that runs over the PCIe interface. Diverge has a SM951 which is a PCIe M.2 drive, both available in AHCI or NVMe. But both types of SSD will be able to work on 4 lanes regardless of it being AHCI or NVMe.
The same is true for the motherboard: AHCI or NVMe won't affect the amount of free PCIe lanes after installing an M.2 SSD.

Why ? Because of the OSI model.
https://en.wikipedia.org/wiki/OSI_model
http://v-integration.com/pcie_primer.htm

Basically: the layer-within-layer principle allows interoperability. So if tomorrow someone brings out some hypothetical NVMe v2.0, it could still run over PCIe if developed within the specifications. That's why we can use various types of devices over a PCIe interface.

You then replied with:
But correct me if I'm wrong but NVME won't "go via" SATA ( PCH/DMI) and therfore both Diverge's (assuming ACHI) and Ail's posts regarding SATA arn't applicable in the discussion earlier where a NVME = PCIE was assumed.
NVMe won't go over SATA but that doesn't matter in relation to which ports or lanes are disabled or not allowed to be used.
Using an M.2 SSD, being SATA (AHCI) or PCIe (NVMe or AHCI) can be responsible for disabling SATA ports to free PCIe lanes on the PCH, in all three those scenarios.
 
Last edited:
Please be clearer when you try to refute someone's argument, I have no idea what you are referring to what I might be wrong about.

I'm not refuting anything, I think you haven't properly read or understood what we talked about.. I'm just watching you going of on a tangent and trying to understand were...

Just to recap: What happens to the PCIe x16 (on a mitx Z170 board) slot when you install a M2 X4 SSD (NVMe), was the topic...

AHCI or NVMe don't decide or influence if the PCIe x16 slot gets to keep all 16 lanes since it's a protocol that runs over the PCIe interface. Diverge has a SM951 which is a PCIe M.2 drive, both available in AHCI or NVMe. But both will be able to work on PCIe 3.0 x4 regardless of it being AHCI or NVMe and the same is true for the motherboard: it won't affect the amount of free PCIe lanes.

My point beeing a AHCI can use SATA Express but NVMe can't therfore, we can't be sure that Diverge has a NVMe and it's unlikely he/she does, since most boards seem to drop the PCIe slot to x8 when Diverges showed it was still running @ x16.
And as a consequence her/his setup is not a good example in that topic...

And then you started your bit..
 
From my understanding M.2 SSDs come in 2 interface flavors; they either use PCIe (AHCI, NVMe - protocols), or SATA.

On the Asrock Z170 Gaming ITX, if you use a SATA M.2 device, it disables 2 SATA ports (as per the manual). It doesn't mention disabling any functions if you use a PCIe M.2, and I'm not seeing my main x16 slot being reduced. Unless the manual is incorrect, and they disable 2 sata ports for both types of M.2 devices.
 
My point beeing a AHCI can use SATA Express but NVMe can't therfore, we can't be sure that Diverge has a NVMe and it's unlikely he/she does, since most boards seem to drop the PCIe slot to x8 when Diverges showed it was still running @ x16.
And as a consequence her/his setup is not a good example in that topic...

And then you started your bit..
Let me ake it clear for you: It makes NO DIFFERENCE WHATSOEVER if your drive is using AHCI or NVMe. Both are command sets that can pass over a bus. There is nothing physically preventing passing NVMe commands over a regular SATA bus, simply that nobody will ever bother to make a controller that does that.
SATA Express is simply a 2x PCIe link. m.2 is a 4x PCIe link. u.2 (though that only defines the device end, not the board end) is a 4x PCIe link. All can connect to drives that use either the AHCI or NVMe commandsets.

Whether a board connects the m.2 slot to PCIe lanes from the PCH or the CPU is dependant on that specific manufacturer, and varies on a board-to-board basis. Which PCIe lanes from the PCH are used for the m.2 port also varies, and the chart I posted shows the possible configurations a motherboard manufacturer can choose. Remember that only PCH PCIe ports 15-26 can be used for RST, and that there will be 2x (but more likely 4x, because it makes splitting easier) occupied by the wireless module. If a thunderbolt controller is used, that's another 4x lanes gone. If you have dual-ethernet on the back panel, that also imposes limitations.
 
Let me ake it clear for you: It makes NO DIFFERENCE WHATSOEVER if your drive is using AHCI or NVMe. yada yada yada

Everybody is so nice and friendly today.. and they sure read back a couple of pages to know what we are talking about...

simply that nobody will ever bother to make a controller that does that

Good, proves my point..

Whether a board connects the m.2 slot to PCIe lanes from the PCH or the CPU is dependant on that specific manufacturer,

Even better, that's exactly my point... Just because it works with Diverges Asrock board doesn't mean it works with the coming Asus board in the discussion..

Also have you meet Phuncz? you both seem to share in interest in theoretical discussions about these things.. (also passive aggressiveness and a incapability to read... ) :rolleyes:
 
We don't know eachother but EdZ seems to share my frustration with you not reading what we are saying. You have some completely wrong statements and we're trying to explain this to you why they are wrong.
 
We don't know eachother but EdZ seems to share my frustration with you not reading what we are saying. You have some completely wrong statements and we're trying to explain this to you why they are wrong.

No, you two are trying to "explain" what theoretically works and why...
The rest of us were talking about a specific circumstance with what's practically doable today and so on...

You're talking about how and why the space shuttles theoretically flies, while the rest of us are sitting thinking about the paper plane we just folded from a piece of paper..
The paper plane "could" probably do stuff which are really cool, but requires a couple of cern's.. and since i don't have even one, i could care less what it "can" do, only what it does here and now..

Two completely different subjects, who might seem similar, but aren't..

That's why you don't think i'm reading what your saying, It's because you're talking about something else..

"There is nothing physically preventing passing NVMe commands over a regular SATA bus, simply that nobody will ever bother to make a controller that does that"
It works in theory, but not in practice...
hence ME:won't work..(implied if you tried it with existing things today) YOU: it's so does, because if you built this and invented this...
 
Last edited:
My point beeing a AHCI can use SATA Express but NVMe can't therfore, we can't be sure that Diverge has a NVMe and it's unlikely he/she does, since most boards seem to drop the PCIe slot to x8 when Diverges showed it was still running @ x16.
And as a consequence her/his setup is not a good example in that topic...

And then you started your bit..


It doesn't matter if my SM951 was the AHCI version, or the NVMe, they are both PCIe devices.

The protocol they talk in doesn't matter, it comes down to if they are a PCIe device or a SATA device... at least for the Asrock Z170 ITX board. If I had a SATA M.2 device, SATA ports get disabled. PCIe M.2 is not documented for having anything disabled, and it doesn't seem to affect my 16x PCIe slot. So until someone does more tests (not it - my build is already in my ncase case), we won't know for this particular board.

edit: M.2 PCIe device are rated for 4x as well, so I don't see any reason why NVMe would be any different
 
So until someone does more tests (not it - my build is already in my ncase case), we won't know for this particular board.

Well, you replied with "works for me on my asrock" (with out providing your exact setup) when JJ from Asus "said in the video" it wouldn't for the specific board we talked about.

We should be able to conclude that the boards work i different ways and what works on your board is not directly transferable to the Asus. (not surprising) And given youre at that time unknown hardware it's more of a stretch to draw any conclusions.

Either you just wanted to say that it works with your combination, and i'll reply, good for you....
Or you tried to draw a similarity between the boards (ie "if it works for me then it maybe works for you") which we now can conclude isn't possible.
Which is what i tried to point out, when asshat 1 and 2 went to town...
 
Calling people asshats because they are clearing up misinformation is not the way to go about things. Whether you, they, or both parties misinterpreted things is irrelevant at this point. Since there is still a risk people will read this thread and be utterly confused, lets recap everything to make sure we are on the right page to prevent the further spread of misinformation.

Sata drives use PCI-e lanes. x1 PCI-E 3.0.
Sata express drives use PCI-e lanes. 2x PCI-E 3.0.
M.2 drives use PCI-e lanes. The fastest use 4x PCI-E 3.0.
U.2 drives use PCI-e lanes. The fastest use 4x PCI-E 3.0.

So yes, all use PCI-E lanes. NVME and AHCI are controller methods. NVME is more efficient for devices that utilize multiple PCI-E lanes, as it has less latency and overhead. AHCI was designed for slower, mechanical drives.

Lets consider one specific scenario:

Samsung 951 AHCI
Samsung 951 NVME

Both use M.2 x4 PCI-E 3.0 for connectivity. There is in fact a specific M.2 socket type for this level of speed. Socket 3. The key type can very, but we won't get into that for this explanation. The difference between AHCI and NVME is the NVME controller can more efficiently handle communication via PCI-E. These controller types will not affect a manufacturers decision on how to split up the lanes and ports for use on a motherboard.

Skylake as an entire platform has the following lanes available for use:

16x PCI-E 3.0 on CPU
20x PCI-E 3.0 on PCH (Chipset/Southbridge).

These lanes must be utilized to provide all connectivity options. The PCH is usually tasked to handle Ethernet, USB, and all port types (Sata/M.2/U.2/Sata Express). The PCH must still communicate to the CPU, through the DMI. DMI 3.0 on Skylake has about 40 Gb/s of bandwidth total. That means everything connected through the PCH has that total amount of bandwidth to travel through. Considering a fast PCI-E 3.0 x4 drive at full speed saturates 32 Gb/s, this is a concern to some manufacturers.

This is why some manufacturers will design boards to pull lanes directly from the CPU for a M.2/U.2 drive, rather than the PCH - even if sufficient lanes on the PCH are available.

How they decide which devices will disable others to gain sufficient PCI-E lanes for bandwidth is dependent on how they envision the usage scenario the motherboard will undertake. EdZ's photo shows how these divisions occur. . Hopefully this provides sufficient clarification with respect to how everything correlates. :p
 

I like the MSI. If it would have come out before the Gigabyte I probably would have went with it.

Bear in mind this model does not appear to support 80mm M2 SSDs

EDIT:: manual is available now. Checked and it says it supports 4.2 and 6cm length cards so no 80mm good catch.
 
Last edited:
Calling people asshats because they are clearing up misinformation is not the way to go about things.

Agreed!

The difference in this case was that they were spreading it, by not understanding what was disussed..

And when you've misunderstood something you might wanna check that you've got it and not act like the "almighty guardian of the internet" because that will get you called an asshat :eek:

You might wanna add (Check your motherboards manual for what works with you board)...
you know for the people like me, who only care what works with my (future) stuff and not others...;)
 
Well, you replied with "works for me on my asrock" (with out providing your exact setup) when JJ from Asus "said in the video" it wouldn't for the specific board we talked about.

We should be able to conclude that the boards work i different ways and what works on your board is not directly transferable to the Asus. (not surprising) And given youre at that time unknown hardware it's more of a stretch to draw any conclusions.

Either you just wanted to say that it works with your combination, and i'll reply, good for you....
Or you tried to draw a similarity between the boards (ie "if it works for me then it maybe works for you") which we now can conclude isn't possible.
Which is what i tried to point out, when asshat 1 and 2 went to town...

... you seem to assume a lot... I never said anything about asus boards, or anything to lead anyone to believe they all are setup the same. My initial post towards this conversation was to show that it doesn't slow down the x16 slot when you use an M.2 PCIe device on this asrock board, like someone assumed it would, just because asus boards do.

If anyone is coming across as an asshat it's you....considering the tone from the post I just quoted.
 
Ok, so now that we've moved past that...

I guess the decision for me is down to the ASRock and the gigabyte gaming board. They unfortunately each have a few cons and I'm trying to figure which would be the best choice. I'm guessing based on the comments in this thread that others are in the same boat. Unfortunately, the new MSI board looked pretty promising, right up to the point that it didn't support 2280 m.2 devices... and then it got rejected.

Both Boards:
+ 2280 M.2
+ USB 3.1 Type A+C
+ AC WiFi

ASRock:
+ 4 additional USB 3.0 ports on the back panel
+ Intel NIC
+ HDMI 2.0 (not relevant to me, but I think this is a critical item for others)

- no alpine ridge so non-intel wifi with bluetooth 4.0 (as opposed to 4.2, the new spec which has specific hardware requirements for certain features)
- only 2.1 analog outputs (not a big deal for me, I'd just use optical, but could be a deal breaker for others)
- ~$20 more expensive

Gigabyte:
+ Alpine Ridge Intel Wifi w/ Bluetooth 4.2 (supports WIDI; Thunderbolt 3? Should have it with Alpine Ridge, but not clear on this - conflicting messages)
+ Full analog audio outputs on the backpanel
+ $165, not $185
+ believe it has better audio hardware (still uses the ALC1150 codec, not too certain about this item)

- Killer 2200 NIC
- Limited selection of USB ports (4 less than ASRock)
- no HDMI 2.0

Have most folks made a decision on their board? This seems to be an issue of Intel NIC (which will be my likely connection) w/ extra USB ports vs the Alpine Ridge controller and saving $20 bucks (for me, other factors listed above could be an issue for others).

Anyone know if the software for either is appreciably better? Would appreciate anyone else's perspective on this. What are you getting?
 
I went with the gigabyte, hope to have it Thursday. The killer nic is a downer but not deal breaker for me, it will work fine, just won't bother with the software. Really, they both seem like great options, I don't think you can go wrong with either. Fan control has been mentioned as a downside for the gigabyte and the ASRock does have an extra fan header if you care about that. I can't comment on software at this point but will chime in once I get it all setup. The brand and the warm fuzzy of alpine ridge won me over in the end I think. I have no experience with asrock but my old g33 gigabyte board is still running strong after 10+? years so that earns some points with me.
 
Quick information: Asus has a rather big cashback event running (at least in germany, i don't know about the other regions). If you buy the Z170i until Oct 10 you'll get 25€ cashback and a code for BF hardline.

Considering the price of only €130 then i'll go for the Asus board. ~60€ more for the ASRock is not worth it IMHO.
 
Quick information: Asus has a rather big cashback event running (at least in germany, i don't know about the other regions). If you buy the Z170i until Oct 10 you'll get 25€ cashback and a code for BF hardline.

Considering the price of only €130 then i'll go for the Asus board. ~60€ more for the ASRock is not worth it IMHO.

I'm living in Spain but could you tell me where I can see this offer please?
 
I'm starting to suspect that Asus video may be incorrect. The Z170i is not listed on Asus' website yet, but their H170i ITX board with a very similar board layout uses the PCH lanes for the m.2 (no mention of the PCIe slot dropping down to 8x, and lists the m.2 as supporting Intel RST, which would not be possible with CPU lanes). Plus, using lanes switched from the CPU to the m.2 slot would require a PCIe lane switcher chip, which is not visible on the board (one couild be on the underside, but this would mean routing the PCIe traces from the front-of-board CPU socket, to the rear-of-board lane switching chip, then back to the front of board again to the slot!) and would add extra cost.

We'll see when the actual documentation is released, but I suspect this was a chunk of script copy & pasted from a video on one of the older mATX/ATX boards where the m.2 slots often were using CPU lanes, because the PCH could not supply PCIe 3.0 in the first place (prior to the 170 series or 99 series chipsets) without someone checking it over first.
 
Yes, it seems like the guy in the video has no idea whatsoever about the board. He also mentioned that the red USB ports are just red USB 3.0 ports, while you can clearly see the asm1142 controller on this hi-res pic ( http://i.imgur.com/yEPWlKd.jpg , behind the red back connector) and the USB 3.1 label on the backplate ( http://geizhals.at/p/6615/1306615/1306615-9.jpg ). So maybe there's still hope that m.2 is in fact connected to pch, but I guess we'll have to wait until the product page goes live... (also weird they got those running for H170 and the Z10 ATX pro gaming but not for the Z170i)
 
I'm leaning toward the ASRock board, but given the confusing litany of posts on this issue, I hope you don't mind if I ask for simple clarification/summation.

I'm building a server (NCASE M1 V4) and don't plan to use the PCI-E slot at all, but having access to the full SATA array is important.

On the ASRock, if I use a PCI-E SSD in the M.2 slot, will that disable any of the SATA ports? I don't care if it slows down (or even entirely disables) the x16 slot, since I won't be using it.

If the answer to that is "yes", does using a PCE-E SSD in the M.2 slot on the Gigabyte disable any of the SATA ports?

I know this has been asked an answered but there is a lot of debate and it was hard for me to filter among all the "I'm not an asshat, you're an asshat" comments.

I really like the ASRock board given the Intel NIC and extra fan header. Plus my X79 ASRock board is still going strong, so my robust sample size of one makes me feel OK about the brand.
 
Ok, so now that we've moved past that...

I guess the decision for me is down to the ASRock and the gigabyte gaming board. They unfortunately each have a few cons and I'm trying to figure which would be the best choice. I'm guessing based on the comments in this thread that others are in the same boat. Unfortunately, the new MSI board looked pretty promising, right up to the point that it didn't support 2280 m.2 devices... and then it got rejected.

Both Boards:
+ 2280 M.2
+ USB 3.1 Type A+C
+ AC WiFi

ASRock:
+ 4 additional USB 3.0 ports on the back panel
+ Intel NIC
+ HDMI 2.0 (not relevant to me, but I think this is a critical item for others)

- no alpine ridge so non-intel wifi with bluetooth 4.0 (as opposed to 4.2, the new spec which has specific hardware requirements for certain features)
- only 2.1 analog outputs (not a big deal for me, I'd just use optical, but could be a deal breaker for others)
- ~$20 more expensive

Gigabyte:
+ Alpine Ridge Intel Wifi w/ Bluetooth 4.2 (supports WIDI; Thunderbolt 3? Should have it with Alpine Ridge, but not clear on this - conflicting messages)
+ Full analog audio outputs on the backpanel
+ $165, not $185
+ believe it has better audio hardware (still uses the ALC1150 codec, not too certain about this item)

- Killer 2200 NIC
- Limited selection of USB ports (4 less than ASRock)
- no HDMI 2.0

Have most folks made a decision on their board? This seems to be an issue of Intel NIC (which will be my likely connection) w/ extra USB ports vs the Alpine Ridge controller and saving $20 bucks (for me, other factors listed above could be an issue for others).

Anyone know if the software for either is appreciably better? Would appreciate anyone else's perspective on this. What are you getting?

Sorry to ask this but what is wrong with the non Gaming Gigabyte?
 
The ASRock Fatal1ty Z170 Gaming-ITX/ac manual only mentions disabling SATA 4 & 5 if a SATA m.2 card is installed. Presumably if a PCIe m.2 drive is used these ports are unaffected (as the SATA warning is prominently repeated, but PCIe m.2 drives are not mentioned).
 
Sorry to ask this but what is wrong with the non Gaming Gigabyte?

it lacks the usb 3.1 and the "wonderful" gaming design. apart from that it is nearly the same, it even has a similar pcb. In my opinion the gaming board is not worth the extra price but I won't get gigabyte because i already have a H77 one and the fan control is horrific, and it hasn't changed since than, however i am satisfied by Gigabyte boards in all other aspects so I'd recommend them if you don't need absolute super-silence :p
 
I went with the asrock one due to availability mostly but honestly since I was going for ultra silent build I'm relieved that the bios fan control is amazing. Worth the $20 extra for me. Also ibhave a video card and pcie x16 video card so that was also a plus. For the gigabyte it was confirmed that it will not have thunderboard 3 even though the manual says it has it. If fan control and m.2 /pcie x16 matters less go for the gigabyte its also a good board.
 
it lacks the usb 3.1 and the "wonderful" gaming design. apart from that it is nearly the same, it even has a similar pcb. In my opinion the gaming board is not worth the extra price but I won't get gigabyte because i already have a H77 one and the fan control is horrific, and it hasn't changed since than, however i am satisfied by Gigabyte boards in all other aspects so I'd recommend them if you don't need absolute super-silence :p

Thanks, I sure can live without USB 3.1. Will the 4 pin ATX power connector have any negative OC effect vs the 8 pin what is being used on the Gaming 5 board?
 
Sorry to ask this but what is wrong with the non Gaming Gigabyte?
Good question, in all honesty I hadn't looked to hard at the WiFi board because I thought that without USB3.1 you wouldn't have the Type C connector. In looking at the website, obviously that isn't the case (thanks for asking the question!). I build a comp once every 3-5 years, and I was planning on building this one out to last a while so I wanted to make it as "future" proof as possible (fools errand :)), so I figured if I could get the new standards that would be ideal since expansion cards obviously aren't going to be much of an option...

In losing 3.1, you really lost the 10 Gbps transfer and the increased power delivery, right? I like some of the things you can do with 3.1 USB Type C (e.g. have a single cable connecting my computer to my monitor delivery power and keyboard / mouse connections, so swapping between my desktop to a laptop is a seamless experience). Would you potentially not be able to use certain devices with just the USB 3.0 over Type C?

I don't care at all about the gaming highlights or heatpipe design, but I do plan to OC the system so not certain if I lose much there. Does anyone have the Gigabyte Wifi? Good experiences out there? Definitely appreciate the discussion in this thread, looking to pull the trigger in the next week or so. Was originally planning to go 6700k, but the 6600k is getting more tempting give that it's over $100 cheaper and the $6700k is so hard to find. I was expecting the pricing to be closer to the 4790k (~$330/$340ish), but it's at $370. Kinda hard to stomach considering you don't get a fan either (which I wouldn't use, but ultimately is a cost reduction for intel).
 
For those interested in the Gigabyte Gaming-5 for the sake of Thunderbolt .......

"Unfortunately Thunderbolt 3 is not supported on this particular board due to the chipset and Bios limitation." from eddie@gigabyte USA

Im not too bothered by it due to the fact I don't feel like $150 more for the Impact (assuming it will have TB3) is worth it for my use. I would rather use that cash for next gen boards.

Still dont get why the User Manual would label "Intel Thunderbolt Controller" instead of 'Alpine Ridge' .......... that is seriously misleading.

Should have my Gaming-5 later today , on its way today for drop-off..
 
Alpine Ridge is the codename for Thunderbolt 3, like Skylake is the codename for the 6th generation Core-i processors. We use it because it's shorter I guess (Skylake is) and most of the public doesn't know Alpine Ridge or Skylake.
 
That still doesn't clarify what the mobo has. It's like the rep saying the 6700k is not supported but an unlocked Skylake i7 is supported in the manual
 
Gigabyte Z170 gaming showed up at my house today. Now for the other bits. The board looks fantastic for what it is worth.
 
That still doesn't clarify what the mobo has. It's like the rep saying the 6700k is not supported but an unlocked Skylake i7 is supported in the manual
It wouldn't be the first time a manual was completely wrong about a feature, I've had plenty of boards with an "addendum" sheet or two stuck in there along the manual, stating new limitations, unsupported features and incompatibility. I would believe the spokesperson before I do a motherboard manual, that is often filled with generic images and text, often not proofread or fact-checked by people qualified.
 
Isn't that always the case? Manual says 'X', manufacturer rep says 'Y', and when the boards actually arrive and people test them it turns out to be 'Z'.
 
it lacks the usb 3.1 and the "wonderful" gaming design. apart from that it is nearly the same, it even has a similar pcb. In my opinion the gaming board is not worth the extra price but I won't get gigabyte because i already have a H77 one and the fan control is horrific, and it hasn't changed since than, however i am satisfied by Gigabyte boards in all other aspects so I'd recommend them if you don't need absolute super-silence :p

The only bad thing about ga-170n-wifi is that its not available yet :)

It should come out in AUS on 19th of September.
 
Board showed up a day early (Gigabyte) so I was able to build last night. At work now so I can only answer some questions but can post pics later on if anyone wants to see. Was only able to get win7 installed before I passed out (need to set legacy mode in bios if installing from disk or the install will freeze) but will get 10 installed tonight and do some testing.

The board does look very nice in person even though I'm not a big red/black fan.

Specs
--------
Gigabyte Z170n - Gaming 5
Intel i5 6600K
16gb (2x8) Corsair Vengance 2666 Ram
Asus GeForce GTX 970 mini
Samsung 850 Evo 500GB SSD
Samsung SM951 M.2 256GB SSD (AHCI)
Random EBay slot load blueray
Silverstone FT-03 mini case
Silverstone SX500-LG SFX-L PS
Noctua NF-P14s
Cryorig C1 ITX Cooler
 
Back
Top