Random build questions

[Muaadib]

Hello,

Recently i decided my old system is a bit long in the tooth, so i decided a small upgrade....then another small upgrade....then another. In the end i found i made many changes but i'm satisfied with my picks. The old system specs is in my profile, I kept the ram and drives from it but the rest changed to:

i5 6600k > r5 2600

Arctic i30 > Arctic Liquid Freezer 120

Gigabyte Z160-HD3P > Asrock B450M Pro4

HiS R9 290 > Sapphire Vega 56 Pulse

CiT Gforce > Coolermaster Q300L

I also bought 2 Arctic (yes i'm a Fan...pun intended) F14 PST for the new build, along with a 50cm 5050 RGB strip and a PCIe TV card. I have a couple of questions in mind that aren't related to each other, just some stuff that's been on my mind for a while:

1. Should i be expecting any improvements in gaming or is this more of a sidegrade? I plan to OC the 2600 to 4.2Ghz, same clock speed i ran the 6600K at. The 2600 will have other productivity uses so i'm not too bothered if it happens to be slower in gaming, just trying to see what to expect.

2. AMD's StoreMI seems lovely to me, but i am very confused on how it applies in my case. I have windows installed on the 250GB SSD and an almost full 2TB HDD. Their guide recommends the following

- Option 1: Erase the HDD and use the existing windows installation on the SSD.

- Option 2: Erase the SSD, install windows on the HDD then add the SSD as a StoreMI drive.

Option 2 is the easier one for me to do as i was planning to reinstall windows anyway, but the thing that is confusing me is: after the SSD is added, will it just extend the Windows partition by 250GB or will it add 250GB to the whole 2TB drive; appearing as 2.25TB now? Is there any other way where i preserve my data on the HDD?

3. After the RGB strip arrived i found out its powered by 5V while the Pro4 seems to have 12V power in its RGB headers. I was thinking i can just connect a 5V pin from a USB 2.0 header to the strip instead. Is this a good solution or am i missing something?

4. Since this will be my first time using an RGB strip, where do people normally place them inside the case? i'm leaning towards Top and Front sides (maybe cut the cable into 2x25cm strips if length is an issue).

5. Where should the 120 AIO be installed? Top of Front / Back?

6. Related to the previous question, If i go with AIO in the front should i use one of the F14's on the top over the VRM pushing/pulling? i read that it's not needed for a 6 core with a heatsink'ed VRM.

7. Bit of a pet peeve, but is there anyway to change the LED color of the Vega 56 Pulse to white or to disable it? I mistakenly thought it had RGB and seems expecting a gpu named "Pulse" to have a pulsating LED is a bit of a far fetch.

8. If the LED cant be changed then i will go for a red build, in which case i will have the choice of using a red Bitfenix Spectre or a much better airflow (unscientific measurement by hand test) white Corsair AF120. Since i value airflow more i would love to use the Corsair but then i would have to find a way to make it glow red, should i paint it? red tint sticker? change LED? any suggestions?


- Some thoughts on the Q300L: The case supports 140mm fans on all 3 sides (top,front,bottom) since the 3 panels are basically the same. I mention this because CoolerMaster's site says it only supports 140mm on the front. Also, I dunno if this is caused by my last decent case being a 2009 purchase (CM Storm Scout) or the fact that i'm coming from an El Cheapo case, but this case impressed me on all fronts, from the dust filters to the perforation to the finish.... $40 buys a whole lot these days.





Thanks for reading the ramblings of a meticulous, slightly ocd dude. Let me know what you think.

 

[SniperHF]

It's sidegradeish, except the Vega which is a solid upgrade on the R9.
The extra threads could help depending on what you're doing, but not much in gaming outside of a few select titles.


I wouldn't mess with StoreMI but I'm a curmudgeon that way. Mainly because I wouldn't want to deal with it down the line if I decided to go back to Intel.

 

[Muaadib]

Thanks for the input, i should have specified the sidegrade question is about the CPU, obviously the vega would be better .

Indeed, the extra threads will come in very handy for my simulation/design workloads.

I dont plan to go back to the blue team anytime soon so hopefully StoreMI will work well for me.

 

[robot zombie]

Sorry if I ramble a bit. I too am a bit OCD.

2600 is respectable in gaming. Really comparable to your current CPU... ...but it is much better at workstation and multithreaded tasks - you'll see pretty big gains there. In games that use 6+ cores/threads, you may notice a difference. I frequently see 6-8 being well-utilized in games, even older ones. Windows 10 also makes good use of those threads. Really hard to say on the gaming, honestly. Generally speaking I wouldn't expect miracles. It's a safe bet that it at least won't be slower. And with the right RAM, it could beat out your old CPU.

2600 is a bangin CPU, for most things - easily the best midrange all-rounder on the market right now imo. Big picture, it is definitely an upgrade. Enjoy it Even though the Vega 56 is more gonna be the heavy hitter in your gaming situation, but you already know that lol

I can chime in a bit on overclocking the 2600. Airflow, settings, limits and such. Been playing with it a lot recently.

I have my 2600 pushed to 4.25GHz @ 1.32v. Finding this to be about the best I can do with my case/airflow setup - past that, voltage needs and temps go wayy wayy up and my cooling can't take it. And honestly, browsing around, I got pretty lucky on voltage. A lot of people need voltages I couldn't begin to run to get that speed stable. Running heavy stress tests, it caps at 80C, though obviously everyday loads are much lower. Temperature kills stability with the Ryzens. Whatever point you hit ~75C is probably gonna be roughly the point past which you'll start crashing and spitting out errors under prolonged workloads. I am running two Corsair ML140's as front intake and the two stock NZXT 120's as top/rear exhaust. CPU cooler is a Scythe Mugen MAX, a fairly bulky 140mm cooler - comparable to a Cryorig H5. Also bear in mind that the x370 board I'm running has probably the best VRM of any board in it's class - without that, I might not have managed. 4.2 isn't all that easy for higher thread Ryzens. Difficulty goes way up with each 100mhz from stock boost. 4.1 is more realistic, and still very good imo.

I mean, any 2600 should manage 4.2, but more often than not the CPU isn't the limitation at that level. I think in your cooling situation is gonna be somewhat of an issue. That little 120mm AIO is going to struggle at least a little to keep the 2600 cool at peak loads on an overclock like that, imo. That's a single rad, right? Push/pull might help a little, for sure. mATX is an issue, too. The 2600 really heats up when overclocked due to the higher core count and infinity fabric. Past 4ghz, there's a massive power demand increase. They really can get quite hot, asking for as much as 170w at my settings.

Best practice with AIO's is to run them as intake, pulling in cool, outside air to help cool the CPU. Absolutely would not have it as an exhaust if you intend to overclock that CPU.

I would probably make it the front intake with another fan complimenting it and make the top/rear exhausts - provided both front fans there are powerful enough. The air moving from the front out to the top/rear will be enough to cool your VRM's, so long as intake is sufficient. In such a small case, making one of those top/rear spots intakes may actually cause turbulence where the two perpendicular paths meet and can hurt temperatures in that region. Two exhausts sucking air right out of that area unimpeded is generally better than pushing the hot air across some convoluted path out. An intake next to an exhaust causes them to fight. The air pushing in is fighting the will of the air trying to be pulled out - it doesn't automatically just push right out through the exhaust. Some pulls away to the rest of the case. The exhaust would have to be significantly stronger.

It's worth a shot but ime it just doesn't work all that well. You could try running two front intakes in addition to the AIO running intake in top or rear and use the remaining spot as exhaust. Sometimes that works well, maybe because the rad chokes the intake and keeps it from fighting the exhaust right next to it. I just don't know if a single exhaust will be enough to draw all of the heat from the CPU out effectively. You'll want to watch VRMs.

Or you could go crazy with all intake on the top/front and put a superpowered fan on the rear for exhaust

Just some ideas. You'll have to play with it, there. Something has me thinking it's gonna be a finnicky situation. Honestly, a 240mm on the front with top/rear exhaust would be more practical.

Nother very important thing with Ryzen. It really, really likes fast RAM. Performance in all areas greatly benefits from tighter ~3200 RAM, minimum. Honestly, it'll be half of the performance increase you see with your overclock. The infinity fabric that the cores use to communicate with each other and cache is tied to RAM clock speed. Its a real bottleneck. I would probably at least take the time to play around OCing your current set. Play with SOC a little to push it up. People recommend 1.2 as a max, but personally I wouldn't go over 1.15. At that point, you'll want to just get RAM that's faster to begin with. The thing about bumping SOC, especially on lower-end mobos, is that the sensors can give you lower than actual voltages. And SOC voltage has a finer margin between safe and not. Easier to push too much current and fry either VRM or CPU without seeing it coming.

Another major limitation with your OC is gonna be your board. Cooling on it looks okay and BIOS is probably sufficient, but that's only a 3-phase doubled VRM on the vcore. All in all the power delivery is lean - not the beefiest components or most reliable design. When pushing max overclocks on upper-level Ryzens, you really want to have at least a quality 4-phase. 6 is ideal. Voltage and current demands get up there and a 3-phases is likely going to get hot and start to droop too much to hold higher clocks stable. Ryzen 2 is notoriously droopy at overclocking voltages, even on really good boards. If you're lucky enough to keep voltages down on your particular chip, and that board has decent LLC curves, you may skirt by. But again temperatures might be a real problem. Keep a close eye.

I'd also highly recommend Ryzen Master for working out your OC's. Just really handy and reliable for playing with voltage/clock. Also good for accurate temperature readings. A lot of sensor monitoring software will feed you off temps that read low by as much as 30C! You could think you're only at 60 when really you're at 90 and about to throttle! Hwinfo is good too - use tctl/tdie.

And I'll leave you with this... IME the 2600 is an excellently balanced chip out of the box. Overclocking yields some gains, but honestly, it boosts up to 3.9 at excellent efficiency, especially for a 12t CPU. It's a cool, solid, middle-road performer out of the box. If you can get say 4-4.1 out of it, that's great. Bear in mind that even though stock boost is 3.9, it may not hold that, while an overclock to 4 will always hold it when it needs to - that'll mean more than getting from 4 to 4.2. And remember, if you don't pull that out, focusing on RAM can still give you pretty huge gains and you can still have something that is legitimately a very respectable performer with exceptionally low power consumption running cool in a nice low-footprint case. That'd be my endgame with those parts. Food for thought.

Enjoy your nice new build!

 

[Muaadib]

That's a very informative post that i will probably come back to more than a few times, thanks mate.

The AIO maybe a petite 120mm but it has a thick Rad (40mm iirc) and comes with 2 fans for push/pull out of the box, it also performed well in the reviews that i read (HardOCP) so I'm hoping it punches way over its weight class.

You definitely seem to have hit the jackpot on your 2600, 4.25GHz at 1.32v seems unreal to me. Here's hoping i can achieve something similar.

Since the case can only support 2 fans per side i was planning to do:
- Front : AIO + 140
- Back : 120
- (Optional) Top : 140 near the back

I think leaving the front top empty would avoid the turbulence issues you mentioned. Hopefully i won't need to use the top 140 because i would like to have a slightly positive pressure since dust accumulation is a very big issue where i live.

I thought about getting some new Samsung B-Die rams but i couldn't really justify the cost ($200+) over just using my current 3000Mhz ones, i guess i will just tinker with them to my heart's content (already have Ryzen RAM calculator to start with).

I picked the motherboard after some very extensive research (i think i have watched hours of buildzoid's VRM and AMD motherboard videos). The VRM may be a doubled 3 phase but it's performance is generally at least the same if not better than a 4 phase (at least when comparing temps). The board has the same VRM and heatsink as the B450M K4 Gaming which did really well when coupled with a 4.2Ghz 2600 (Hardware Unboxed). I definitely will be keeping my eye on its temps and may even stick a K-couple probe in there just to be sure.

Definitely gonna be using Ryzen Master, falsely reporting temps by 30 degrees is stuff of nightmares to me.


I will try to keep this post updated once the cpu finally arrives, i'm currently just oogling the case and motherboard, unboxing them once or twice a day and imagining how the build would look and possible issues, then reboxing them and swearing i won't do it again until the stuff arrives. That lasts for a few hours before i do it all over again

 

[robot zombie]

The AIO maybe a petite 120mm but it has a thick Rad (40mm iirc) and comes with 2 fans for push/pull out of the box, it also performed well in the reviews that i read (HardOCP) so I'm hoping it punches way over its weight class.

So I've seen. That rad definitely is beefy. Might just get ya there. It'll be cutting it close, though!

You definitely seem to have hit the jackpot on your 2600, 4.25GHz at 1.32v seems unreal to me. Here's hoping i can achieve something similar.

No doubt, though it took a lot of messing around to get that to work. I'm still hoping to try for 4.3. All I can say is that all of your VRM settings matter a lot. You gotta go beyond just finding the flattest LLC. I think a lot people wind up needing more voltage than they should simply because the voltage curve itself has problems. It almost seems to favor a certain amount of droop. Overshoot is really common and it kills stability. You actually wind up increasing voltage drastically to get around it, as dumb as that sounds. And then sometimes if the curve is too flat, the temperature gets away from you and causes errors/crashes.

A lot of it just isn't intuitive. Sometimes you're at what you think should be the right voltage, but actually the way it fluctuates trips up Ryzens... ...or maybe it's too high and the spikes are actually destabilizing it or raising temps to unstable levels for just long enough. Even if max temp is acceptable, a spike within the upper range can still bring it all down. It's kind of baffling when 1.4 hot doesn't work but 1.35 drooping does, even though neither take you quite into throttling territory, but it can happen easily. Again, how your VRM runs matters in ways that don't always make immediate sense. More annoying still, the VRM performance can change a lot with slight voltage tweaks, meaning you sometimes have to go back and tweak VRM settings alongside of slight voltage adjustments. Up yoiur voltage for stability and suddenly your VRM isn't responding quite right. It's a complicated sweet spot game.

RAM overclocks influence CPU voltage requirements, too - like, massively. You can have a RAM overclock that you've tested as "stable" but you go to up CPU core clock and suddenly you find yourself running absurd voltages for a teeny tiny clock speed increase. This would make you think that you're approaching the clock speed limit of the CPU, but actually it's the RAM overclock destabilizing it and the voltage boost just happens to be carrying it. Meanwhile, your SOC voltage is too high and it's bringing up core temps. Touchy memory controller.

Also really can't understate how much temperature matters. It's not just a safety thing. Even in the safe range, max stable temps are touchy and situational. I've had some where it would just crash every time it passed 80C by much, even though it should technically do it. The higher you clock, the lower the max stable temp. Lower clocks could run at higher voltages and just up above a riskier 90C without crashing or throttling. Signs and wonders...

Point is, I had to play around with a lot of different factors to get that voltage level working at 4.25. I started off needing something like 1.375v for 4.2 to pass any stress test - with better cooling I would've never known, but I was down an intake fan and pushing almost 90C, so I had to dig deeper. I think most people accept the higher voltages because they have the cooling to handle it and it's in the safe range. Doesn't mean there aren't ways to improve efficiency, though. Sometimes increasing voltage isn't solving the real problem, even if it does run stable. Still, it does eventually start ramping up hardcore. For me, that's approaching 4.3, when I start needing .05v more for each 25mhz extra.

If you happen to find yourself needing to drastically increase voltage in order to get a 50-100mhz higher when it seems like it shouldn't be, there's probably something else that needs looking into. And it'll probably be something stupid that makes no sense You can save a lot of headache by running RAM stock until you get CPU clocks nailed down, too.

Since the case can only support 2 fans per side i was planning to do:
- Front : AIO + 140
- Back : 120
- (Optional) Top : 140 near the back

I think leaving the front top empty would avoid the turbulence issues you mentioned. Hopefully i won't need to use the top 140 because i would like to have a slightly positive pressure since dust accumulation is a very big issue where i live.

Sounds good man! I wouldn't bother with the front top either. That's just at a point where you're trying to move more air than the space inside the case can accommodate - just too much going on. I hear you on positive pressure, but there are ways around that. You could put weaker fans for exhaust than you have for intake. Take my case. Two high performance static pressure 140's in the front, and two average "airflow" fans in the back. That way even if they all max it's still balanced for slight + pressure, but all in all I can move decent amounts out of the case.

It's probably best to get as close to neutral as possible without going to it or into negative. That way you're moving the maximum amount of hot air out of the case without losing dust rejection. It can only cool as quickly as it can ditch the hot air. I dunno, that seems to work best for me.

I thought about getting some new Samsung B-Die rams but i couldn't really justify the cost ($200+) over just using my current 3000Mhz ones, i guess i will just tinker with them to my heart's content (already have Ryzen RAM calculator to start with).

Haha that's what I've been doing too. I took my old 2400 up to 3000 with an SOC boost and a slight timing tick from 16-16-16 to 17-18-18. Still for reasons mentioned above I feel it is holding me back. I'm gonna ultimately wind up biting the bullet. SOC demands and decreased stability keeps me from upping clock/voltage. And downclock honestly isn't worth it. 2666 doesn't bench nearly as high as 3000. It's actually quite shocking how much of a difference it makes.

I picked the motherboard after some very extensive research (i think i have watched hours of buildzoid's VRM and AMD motherboard videos). The VRM may be a doubled 3 phase but it's performance is generally at least the same if not better than a 4 phase (at least when comparing temps). The board has the same VRM and heatsink as the B450M K4 Gaming which did really well when coupled with a 4.2Ghz 2600 (Hardware Unboxed). I definitely will be keeping my eye on its temps and may even stick a K-couple probe in there just to be sure.

Sounds like you did your homework. I generally trust the guy with the hair. Him and his videos on that stuff are a mess, but there's a lot of good stuff in there. Very interesting that temps are comparable. But remember it's not all temps. Voltage/current consistency matters, too. More so with Ryzen than any other CPU I've worked with.

Definitely gonna be using Ryzen Master, falsely reporting temps by 30 degrees is stuff of nightmares to me.

Haha for real, it's crazy. With certain Ryzen chips monitoring programs report socket temperatures instead of actual temperatures inside of the core. There was also this weird intentional offset with I believe the 1700x and 1800x. For some tctl had a 20C offset and tdie was what you needed. Some, like the 2600 have tctl/tdie combined and that's the correct one. With my motherboard "CPU" temperature is actually off near the socket and that's the reading my fan software applies fan curves to. I remember watching actual temps climb to 90 while that "CPU" sensor was reading 62, and that 62 is what most software would report, while Ryzen master showed the actual.

I will try to keep this post updated once the cpu finally arrives, i'm currently just oogling the case and motherboard, unboxing them once or twice a day and imagining how the build would look and possible issues, then reboxing them and swearing i won't do it again until the stuff arrives. That lasts for a few hours before i do it all over again

Look forward to seeing how it all goes for you. And man, I know that feeling. I was there in January. It's maddening when you're looking at a case, mobo, and PSU, trying not to put it together because you're missing the SSD and CPU

 

[Muaadib]

So I've seen. That rad definitely is beefy. Might just get ya there. It'll be cutting it close, though!

Upon further inspection, it seems its not 40 but a whopping 49mm! really happy with it since i got the AIO for about 55$.

Point is, I had to play around with a lot of different factors to get that voltage level working at 4.25. I started off needing something like 1.375v for 4.2 to pass any stress test - with better cooling I would've never known, but I was down an intake fan and pushing almost 90C, so I had to dig deeper. I think most people accept the higher voltages because they have the cooling to handle it and it's in the safe range. Doesn't mean there aren't ways to improve efficiency, though. Sometimes increasing voltage isn't solving the real problem, even if it does run stable. Still, it does eventually start ramping up hardcore. For me, that's approaching 4.3, when I start needing .05v more for each 25mhz extra.

I love to tinker so i will be definitely changing dials around here and there for a couple of months.

If you happen to find yourself needing to drastically increase voltage in order to get a 50-100mhz higher when it seems like it shouldn't be, there's probably something else that needs looking into. And it'll probably be something stupid that makes no sense You can save a lot of headache by running RAM stock until you get CPU clocks nailed down, too.

Duly noted, gonna focus on the cpu first anyway since my workloads arent that RAM intensive.

Sounds good man! I wouldn't bother with the front top either. That's just at a point where you're trying to move more air than the space inside the case can accommodate - just too much going on. I hear you on positive pressure, but there are ways around that. You could put weaker fans for exhaust than you have for intake. Take my case. Two high performance static pressure 140's in the front, and two average "airflow" fans in the back. That way even if they all max it's still balanced for slight + pressure, but all in all I can move decent amounts out of the case.

I think you mean the Top back? the front will be populated by the 140 in the bottom to supply the gpu with air and the AIO in the top. Sadly i don't have access to static pressure fans so i will have to make due with the Arctics. They arent particularly good in either AF or SP but they bring a good balance inbetween, they are incredibly quiet as well which is a must for me.

Sounds like you did your homework. I generally trust the guy with the hair. Him and his videos on that stuff are a mess, but there's a lot of good stuff in there.

I know what you mean when you say his stuff is a mess, the guy says "So......yea" almost a hundred times per video but he still offers useful information to consider.

I remember watching actual temps climb to 90 while that "CPU" sensor was reading 62, and that 62 is what most software would report, while Ryzen master showed the actual.

I personally go with the totally quick and unsafe way of touching the heatsinks to confirm any suspicious temps. I found a couple of faulty sensors in an old motherboard this way (reported 45c while it really was 70-80s). Thankfully no burnt hands so far. I will use a K-couple probe going forward since i will be receiving a multimeter and some soldering tools with the CPU.

Also, thanks for making me realize that there is a multi quote button

 

[robot zombie]

Duly noted, gonna focus on the cpu first anyway since my workloads arent that RAM intensive.

It's not just about that. The CPU itself runs faster with faster ram. Basically increases the rate at which IC's can communicate. Think of it like this... ...with Ryzen, RAM speed is like FSB speed, or HT speed. So whether you need fast ram directly or not, performance in just about everything benefits massively from it.

I think you mean the Top back? the front will be populated by the 140 in the bottom to supply the gpu with air and the AIO in the top. Sadly i don't have access to static pressure fans so i will have to make due with the Arctics. They arent particularly good in either AF or SP but they bring a good balance inbetween, they are incredibly quiet as well which is a must for me.

These days I don't think SP matters so much, even with rads. They just tend to be the more powerful fans these days. Was just giving an example. And when I said that, I mean the front spot of the top, not the top spot of the front

I know what you mean when you say his stuff is a mess, the guy says "So......yea" almost a hundred times per video but he still offers useful information to consider.

I think he's a pretty smart guy, if you pay attention you pick up a lot of little things that come in handy. I just think it's funny how he rambles. I do the same crap, though, heh.

I personally go with the totally quick and unsafe way of touching the heatsinks to confirm any suspicious temps. I found a couple of faulty sensors in an old motherboard this way (reported 45c while it really was 70-80s). Thankfully no burnt hands so far. I will use a K-couple probe going forward since i will be receiving a multimeter and some soldering tools with the CPU.

Bahah I do the same thing. I'll also say it's really annoying how few boards have built in VRM temperature sensors. Lately I've been using an infrared thermometer to get in the ballpark. I'd like to go the thermocouple route myself. That or shove compatible sensor on the back and hook it up the temperature probe pins on my mobo.

Also, thanks for making me realize that there is a multi quote button

LMAO, you're welcome?

 

[Muaadib]

Weekend update: Got my Vega 56, proceeded to play around with it, then thought about flashing the "balanced" BIOS to vega64 to unlock more HBM voltage headroom. I tried to flick the BIOS switch but it was too far for my fingers, so i used a screwdriver and to my horror it broke the switch and the card wouldn't post.

After looking online i found some advice to use a needle to push the switch to either position. In another horror episode the needle actually broke off the copper piece inside the switch, rendering it ultimately useless.

After kicking myself for a few hours, calculating how much time (and cost) it would take to RMA the card (Hint, i received the card now even though i bought it mid june) I said "Fuck it, i will fix it myself"

Completely disassembled the card (most likely voiding my warranty), picked away the plastic part of the switch and revealed the 3 pins beneath, mentally imagined how the switch would work and concluded that its Pins 1-2 for the turbo bios, Pins 2-3 for the balanced one.

I soldered Pins 1-2 and booted the card as a secondary in my system, atiwinflash recognized it as vega but couldnt read its BIOS. Fortunately i already backed all them up so after a quick reflash to the turbo one later and very nervous seconds between pressing the power button and looking at the screen.... it worked flawlessly.


Granted, temps are a few degrees higher now, possibly due to me not having TIM to reapply but i will do so in the future, i also plan to use a 3-prong switch to make a DIY Bios switch so i can have access to the 2nd bios without having to resolder stuff. I will make sure it sticks out well next time

 

[Splinterdog]

To the OP
Did you try using StoreMI as per your first post here? I'm looking at it, but I'm not sure whether to move data away from the SSD/HDD first and start with a blank sheet. Or do that and move some games onto the HDD, then install StoreMI.
I have a similar setup to you,so I'm curious about your experience with this.

 

[Muaadib]

Sorry i've been busy with some stuff in the last few weeks. Yes i did try StoreMI but i ran into some unrelated ram BSOD issues. I mistakenly thought the issues were caused by StoreMI so i removed it but i haven't had time to go back to it yet.

The first thing you should do is backup everything, once you install StoreMI you cant return to how your system is setup without wiping both drives. The easiest way to do it imo is install Windows in the HDD (slow tier). You can partition the HDD however you like before you install StoreMI as it wont wipe the drive. Afterwards you can add the blank SSD as a fast tier to your existing HDD.

P.S. One weird issue i faced while using StoreMI was corrupt downloads. They would finish 100% but i would get a CRC error when running any downloaded executable file. It went away when i stopped using StoreMI. I still have no clue what caused that.


Back to my own post, since i did the build (minus the AIO as it hasnt arrived yet) i will answer my own questions as it may help others.

1. Should i be expecting any improvements in gaming or is this more of a sidegrade? I plan to OC the 2600 to 4.2Ghz, same clock speed i ran the 6600K at. The 2600 will have other productivity uses so i'm not too bothered if it happens to be slower in gaming, just trying to see what to expect.

Performance seems to be about the same, i haven't had time to OC yet but in general it feels like a sidegrade

2. AMD's StoreMI seems lovely to me, but i am very confused on how it applies in my case. I have windows installed on the 250GB SSD and an almost full 2TB HDD. Their guide recommends the following

- Option 1: Erase the HDD and use the existing windows installation on the SSD.

- Option 2: Erase the SSD, install windows on the HDD then add the SSD as a StoreMI drive.

Option 2 is the easier one for me to do as i was planning to reinstall windows anyway, but the thing that is confusing me is: after the SSD is added, will it just extend the Windows partition by 250GB or will it add 250GB to the whole 2TB drive; appearing as 2.25TB now? Is there any other way where i preserve my data on the HDD?

After the SSD is added, it shows as extra space on the C: drive, or as unallocated space that you can partition yourself.

3. After the RGB strip arrived i found out its powered by 5V while the Pro4 seems to have 12V power in its RGB headers. I was thinking i can just connect a 5V pin from a USB 2.0 header to the strip instead. Is this a good solution or am i missing something?

Using the USB2.0 headers worked fine, had to mess around with soldering for a bit though on the actual cables.

4. Since this will be my first time using an RGB strip, where do people normally place them inside the case? i'm leaning towards Top and Front sides (maybe cut the cable into 2x25cm strips if length is an issue).

I went with top and front, seemed the logical path since i moved the front IO to the bottom where it hides my PSU cables.

5. Where should the 120 AIO be installed? Top of Front / Back?

No AIO yet, though i decided on Front mounting it once it arrives.

6. Related to the previous question, If i go with AIO in the front should i use one of the F14's on the top over the VRM pushing/pulling? i read that it's not needed for a 6 core with a heatsink'ed VRM.

VRM temps seems very normal so far, will have to properly test when i get to OC'ing the pc.

7. Bit of a pet peeve, but is there anyway to change the LED color of the Vega 56 Pulse to white or to disable it? I mistakenly thought it had RGB and seems expecting a gpu named "Pulse" to have a pulsating LED is a bit of a far fetch.

Couldn't find a way to change the color so i just unplugged the Pulse logo LED when i was fixing the broken card. Might plug it back in later if i decide to go with a red theme.

8. If the LED cant be changed then i will go for a red build, in which case i will have the choice of using a red Bitfenix Spectre or a much better airflow (unscientific measurement by hand test) white Corsair AF120. Since i value airflow more i would love to use the Corsair but then i would have to find a way to make it glow red, should i paint it? red tint sticker? change LED? any suggestions?

Probably gonna swap a white AF120 for a friend's SP120 (Non led, red halo).