im actually looking for newer parts and priority on durability. Hence 580 > 480, as you said newer pcb mean better build? asrock is the only mobo manufacturer emphasize so much on their durability ( better pcb, premium choke, nichicon caps) *maybe i fell for their marketing? im not sure* and its one of the the cheapest x370 board around.

new pcb, newer vvrm, better cooling, and 8+6pin power connector? XD
I had watch youtube which breakdown taichi, the vrm design and component model is quiet 1 of best higher end x370.
As audio part, asrock also done a well job as well, its worth to get it if willing to pay for better X370.

Asus got a nearly same vrm as well, but i didnt check its breakdown, and i dont really like asus product, from my past experience.

MSI X370 titanium is using the vrm which u can find even at B350 mobo...

aaah! i wondered why their b350 is the most expensive one. now thats explained it. i heard a lot good things about b350 tomahawk but just topup a bit can get x370 from asrock.

Any one here CF a Powercolor pcs+ 390 here? How's the thermals like and what casing you are using?

This post has been edited by ryder: Apr 21 2017, 08:25 AM

Becareful asrock mid end x370 board also mix with average vrm components, some using niko-sem and some using sinopower

Am I safe to say that msi b350 tomahawk has better components than asrock x370 gaming k4? But msi with less features?

On paper,
Asrock gaming k4/sli X370 mobo has 4X2+2x2 power phase, IR35201 as power phase controller. PZ0903BK as highside MOSFET, PK618BA as lowside MOSFET. 12k polymer solid Capacitor.

MSI krait X370 has 4x2+2 power phase, RT8894A as power phase controller. PK616BA as highside MOSFET, PK632BA as lowside MOSFET.

MSI tomahawk B350 has 4+2 power phase, RT8894A as power phase controller. PK616BA as highside MOSFET, PK632BA as lowside MOSFET.

MSI carbon pro B350 has 4x2+2 power phase, RT8894A as power phase controller. PK616BA as highside MOSFET, PK632BA as lowside MOSFET.


In comparison,
IR35201 which commonly found in top tier mobo is greater controller than RT8894A .

PZ0903BK is newer gen, there's no datasheet online yet, so is hard comparable to PK616BA.

PK618BA is 5.5 mΩ 59A and PK632BA is 3.3mΩ 78A, both is same range MOSFET as their price is same range. PK632A should better than PK618BA due to higher amps?
Actually, ryzen draw only low power for SOC, commonly is enough for both, only APU need draw more due to iGPU usage.

Doubler of VRM is better for cooling and stability in theory, my opinion, asrock x370 k4 still better than msi carbon pro, even lower end of msi X370. MSI overpriced for above RM500 mobo...

This post has been edited by eDwanD: Apr 21 2017, 12:30 PM

My asrock x370 fatality k4 is using sinopower instead of niko sem vrm, peoples are saying sinopower is better than niko sem (i totally have no idea, i only know Texas Instrument is the best ). When i bought the board i never check all these about . Initially i also plan to buy tomahawk b350 but after knowing that price for asrock x370 boards (killer and fatality k4) not much different from tomahawk i immediately decide to buy fatality k4.

You had remove ur heatsink? As info in net, there r two possiblity mosfet of asrock k4/sli, either sinopower or niko sem.

Yup, removed and seen, the VRM chips has sinopower logo on top. And yup according to online info either sinopower or niko sem. I oc my 1700 to 3.9ghz 1.3v running aida64 stability test the heatsink is pretty hot to touch, too bad the board do not have temperature sensor for vrm therefore unable to check with the exact temperature.

Should be ok with 1.3v @@ I saw pp from reddit post taht he fried mobo x370 (no model stated) when oc... i forget how much vcore he used for, but in below 1.4v is hard to imagine how to fried the mobo as ryzen consume draw lower amp and watt.

Had u tried to 4ghz ? seems u got great silicon which can oc 3.9 at 1.3v

This post has been edited by eDwanD: Apr 21 2017, 01:14 PM

I think is quite normal for 1700 to do 3.9, i'm not sure. But my friend's one not even to run at 3.8 regardless any voltage input. I haven't try 4.0 yet, even though i have watercooled the chip but the temperature seem abit high too, idle ranging from 30+ to 40+, full load (aida64 stability test) around 60+ to 70.

According to the statistic,
Ryzen 7 1700
93% reach 3.8GHz @ 1.376V
70% reach 3.9GHz @ 1.408V
20% reach 4.0GHz @ 1.440V

I had checked OC form who filled under overclock club forum? Get 3.9ghz wit 1.3v is rare, commonly hit up 1.35v and nearlu hit 1.4v. The best silicon lottery, is hit 4.0 and above with 1.3v.

Whats your water cooling model? is it custom? I think u wont use ur pc daily as under strees test?@@

This post has been edited by eDwanD: Apr 21 2017, 01:35 PM

Finally Vega?!

This post has been edited by eDwanD: Apr 21 2017, 01:35 PM

Yup i'm using custom watercooling which was used for my x99 system, one 480mm radiator with 8x120mm fan push pull configuration for the cpu itself. I'm still try to figure out how to cool down the vrm before i can push further, i'm worried that the mobo will fried out since it is not using high quality VRM chips if i increase more the voltage

thank you so much! now i can finally pull the trigger.

I straight pump 1.45vcore for Ryzen, after that of course you have to tune down la.

K4 got vrm temperature sensor, use hwinfo to check.

Is it? I'm not sure which one cause it using some kind of code.

Seems, RX 550 and 560 is the mc of RX5xx series.